Dani Rodrik on Globalization, Development, and Employment
Apr 11 2011

Dani Rodrik of Harvard University talks with EconTalk host Russ Roberts about trade, the labor market, and trade policy. Drawing on a recent paper with Margaret McMillan on trade and productivity, Rodrik argues that countries have very differing abilities to respond to increases in productivity that allow production to expand using fewer workers in a particular sector. When workers are displaced by productivity increases, what is their next best alternative? Rodrik discusses how this varies across countries and policies that might improve matters. He argues that poor countries should subsidize new products as a way of overcoming uncertainty and externalities from new ventures.

Paul Collier on the Bottom Billion
Paul Collier of Oxford University talks about the ideas in his recent book, The Bottom Billion, an analysis of why the poorest countries in the world fail to grow. He talks about conflict, natural resources, being landlocked, and bad governance,...
William Easterly on Growth, Poverty, and Aid
William Easterly of NYU talks about why some nations escape poverty while others do not, why aid almost always fails to create growth, and what can realistically be done to help the poorest people in the world.
Explore audio transcript, further reading that will help you delve deeper into this week’s episode, and vigorous conversations in the form of our comments section below.


Mads Lindstrøm
Apr 11 2011 at 1:30pm

The idea of banks operating in Italia living up to Italien rules, banks in USA living up to American rules, … seems really sound to me. I have been thinking along the same lines. I like it for multiple reasons:

1) It puts the control, cost and benefits of strict/lax regulation at the same place.

2) I have a lot more faith in nation states making sound decisions than supernatural bodies like UN and EU.

3) We would get competitive set of rules, and we could see what worked best

But I always assumed that there were some very good reasons for not keeping it at the national level. Maybe the cost of less competition, as a bank in country X, would be discouraged from making a subdivision in country Y, if it had to live up to very different rules.

Are there any other reasons, why we do not have a regime of national rules?

Apr 11 2011 at 1:57pm

@Mads Lindstrøm,

Short answer–lobbying..

Because in the modern day capital is pretty much free to flow globally,it investors try to find the place where returns are the highest. if lax regulation allows foreign banks to take bigger risks than domestic firms, the domestic firms are going to lose market share and lobby the government to relax regulations and increase their competitiveness. thus everything ends up at the lowest common denominator, global rules are supposed to prevent this ‘race to the bottom’ however as mr rodrik points out, if the rules are too strict the countries with already lax laws wont agree to them, lest their domestic firms find themselves suddenly uncompetitive.

Apr 11 2011 at 3:18pm

I love the recommendation that currencies should be undervalued–if you’re an elite running a country, sure, go ahead. It grows the overall economy, which you take a cut of, so you get wealthier and look great because the overall economy grows. What few notice is the massive amounts of wealth that you’re depriving your workers because they cannot afford to buy as many imports as they would if your currency was floating and was appreciating over time.

Undervaluing the currency helps the oligarchy running the country and their favorites, while undermining the ability of their workers to grow into a productive middle class. It’s wrong to steal from people, and undervaluing the currency is doing just that. Anyone advocating otherwise has a fundamentally flawed view of the world.

Mads Lindstrøm
Apr 11 2011 at 3:56pm


But if, banks operating in Germany should abide by German rules – foreign and domestic – the difference in regulation would evaporate. The race to the bottom would disappear. Imagine Bank of America wanted to start a bank in Germany, the law could require them to make a subsidiary that abides by German laws.

If Germany had more strict banking rules than America, it might turn some investors off, and German customers would get lower interest for their deposits and pay higher interest for their loans. On the other hand German tax payers would have to pay for fewer bail-outs. My point is that the investors would be there, you just have to attract them somehow.

Apr 11 2011 at 6:40pm

@Mads Lindstrøm,

Imagine Bank of America wanted to start a bank in Germany, the law could require them to make a subsidiary that abides by German laws.

I don’t see what the relevance is, if the german banks had low returns few would invest in them. People would invest in the banks where regulation is more permissive. Investors care about returns and risk, but risk is a distant second, really by necessity. Its very easy to judge returns, but risk is very hard to judge. Nothing to banks did was judged by most people as particularly risky prior to 2008. In fact most of this financial innovation was lauded as making capital markets much safer. Even the big institutional investors (evidently) did not have enough information to determine the risks of these trades.

A great example is the canadian banks, they did not need a bailout. The regulation was better, the corporate governance was better, but the returns were not spectacular. Even after the crisis investment should have flooded out of the US banking sector into the canadian banking sector (if investors are averse to risk) but it did not.

The money will always flow to the least regulated (to an extent) because the returns will always be higher

Apr 12 2011 at 10:23am

Great podcast! I have a question though:

As I understood Rodrik, subsidies are necessary to encourage first investments in a non-existent industry. I understood it in the following way:
Say, an investor would require 100$ in research cost to affirm that a call center would be profitable. He expects income of 100$ and personnel costs of 50$ for the first year, leaving him with 50$ in the red.
Basically at this point he wouldn’t invest, because he would be out 50$ after the first year and in the second year, competition without would drive down income to costs.
If the increased economic activity increases welfare in the society more than the the state would have to tax people, the state should compensate the first investor either with a monopoly for the first two years or a direct subsidy of 50$ if he builds the call center.

Now, did I get that right? What would be the reasoning against that?

Apr 12 2011 at 10:46am

Great, great podcast! I appreciated hearing the perspective of someone who promotes government interventions but seems to, mostly, understand the dangers of the resulting politics. His comments about currency manipulation vs micro-management were very good. I also liked his acknowledgment that he really doesn’t study the US and therefore can’t back up his sense of Keynesian aggregate demand issues.

I did sense that Roberts let Rodrik off too easy on the financial regulation discussion near the end. Rodrik acknowledged all of the problems with international regulations but seemed to be quite comfortable that more national regulations would have good outcomes.

Apr 12 2011 at 9:47pm

This was a great podcast. The best in a long time.

Simon C
Apr 12 2011 at 10:58pm

Rodrik seems to suggest that undervaluing a currency is good for a nation. Presumably this can only be done by the government selling the currency and buying foreign currencies. At some point this will be reversed when the trades are undone and a similar period of overvaluation must occur. Is the whole operation beneficial or just the initial part?

Also, if we consider the welfare of the whole world not just the welfare of individual nations are such currency manipulations still globally beneficial?

Cowboy Prof
Apr 13 2011 at 12:36am


I listened to this podcast while mowing the lawn and had a bit of a hard time following it, but I did pick up on Rodrik’s interventionist bent. It always seemed premised on the fact that we can be “smart” about intervention. The Asian tigers were used as the example of smart intervention, whereas Latin America and Africa not so much. That observation only begs the question of why the Asian tigers were so successful at this.

I think Rodrik’s taste for interventionism is premised on the idea that political leaders will take total social welfare as their principal maximand. While there was a good deal of discussion about economic rents (obtained via first mover status, etc.), he really ignored the question of economic rent seeking through the political process and the subsequent political rent seeking wherein politicians maximize their power rather than social welfare. There was a brief nod to this about 3/4 of the way through the podcast, but Rodrik didn’t seem to take it as a serious concern.

The big problem is that the ability of a political actor to intervene for the “common good” (in an economically rational way that maximizes total social welfare) ALWAYS comes with the ability to seek political rents. Obtaining political rents are frequently at odds with the goal of serving the “common good,” especially if the political actor has a short time horizon.

Over time, I have become increasingly dubious of interventionist claims.

Matt L
Apr 13 2011 at 1:26am

Great podcast. I have an International Finance Master’s and found Rodrik’s research among the most fascinating I came across. Can’t wait for Munger!

Peter C.
Apr 13 2011 at 7:04am

Very good show, especially the difference between South America and China in economics strategy. Could you please have an another economist Ha-Joon Chang (http://en.wikipedia.org/wiki/Ha-Joon_Chang) as your guest? He is currently at Cambridge. His books: “Bad Samaritans” and “23 Things They Don’t Tell You About Capitalism” are must read if you care to know more about modern economics development other than the school of “University of Chicago”. Thanks again for the show.

john berg
Apr 13 2011 at 11:25am

This point is more pro-US constitutional national sovereignty than anti-globalization.

Our US Constitution is being changed, not by explicit Constitutional Amendment, but implicitly by lack of defense by the US Congress–and I mean both houses. A small but significant example: who is responsible for demanding the birth certificate of President Obama. I feel strongly that every member of Congress has the responsibility to require its presentation. The Constitution is written clearly but some have used ridicule to prevent a request for disclosure and now we have the precedent set for ignoring a clause of the Constitution.

As Professor Rodrik notes, the Constitution grants an individual but limited right of monopoly on the issuance of an intellectual property right to an inventor for disclosing his idea publicly. Currently this Constitutional right is about to be “amended” by a mere law that grants this property right to the “first to file” in order to harmonize it with European law.

Nowhere else can one see more clearly the difference between an “individual right” and a “human right.” The Constitution writers saw that only the inventor would, and should, be motivated to invent more by the granting him a property right–-not the entire species.

After approximately 250 years of experience with giving the inventor an ability to profit from his invention, it may never be clear what percentage of our GDP is due to rewarding an inventor for sharing his invention but the US patent system clearly has contributed to our greatness.

John Berg

Mads Lindstrøm
Apr 13 2011 at 12:33pm

@john berg

Maybe I am nitpicking here, but the constitution do not grant patent protection rights to anybody. It grants congress the right to make patent laws, if the laws are for the common good. But congress do not have to use this right.

It is not clear to me, that patents obviously benefits the economy, except for something like the medicine industry, where the time between invention and sales is many years.

The reason I do not find it obvious that patents benefits the economy, is that they incurs huge costs too:

1) Patents attorneys and patent examiners has to get paid. So do people writing patents and judges.

2) When making a product you run the risk of inadvertently braking some patent. This creates risks, which discourages investments. This problem is even greater with lax patent examinations. The patents may not hold up in courts, but companies still incur costs by being sued.

3) Monopoly costs. If something would not have been invented if it were not for patents, then the monopoly costs do not matter, but often things would have been invented anyway. Many companies do not invent because of patents, but if they do invent something they figure they can just as well get a patent. Secondly, many inventions are “invented” multiple times, and even if the first time was driven by patents, it may not be the case that the second time was.

4) It seems dubious to me that government bureaucrats nor judges can reasonably judge if something is an invention, or just a reasonable design to a problem. The most brilliant inventions seems simple when already invented, and many seemingly clever solutions it what any good practitioner would come up with facing the same problem. This is mainly a problem in young industries. In mature industries, the fact that something has not been invented yet, is a clear indication that is not obvious once invented.

I am not claiming that patents is a drag on the economy, but I am suggesting that it is not obvious that it helps the economy either.

Apr 13 2011 at 2:23pm

Great podcast.

Bank regulation idea is intriguing but. As others have noted, practically segmenting bank businesses is very hard. But implementing any of the regulatory fixes is going to be hard and it is far from clear this one is more challenging.

It is notable that Canadian bank shareholders did quite a bit better than their American counterparts.

Great podcast. Thanks to Russ and Dani

Apr 13 2011 at 2:24pm

@Demaratus: Do you think the Chinese people (or Koreans for that matter) have been subject to theft that has hampered the growth of a middle class because their currencies have been undervalued?

Which countries would you point to that have prospered from lack of such theft?

Your analysis is flawed, among other reasons, by assuming the situation is static.

Apr 13 2011 at 4:52pm

@Jonathan: I would argue that a potential Chinese middle class has suffered under the policy of their government which undervalues their currency; certainly the Communist party has its interest in keeping the population of the powerful small enough that it can co-opt them all, so channeling the wealth to the commanding heights of the economy, which allows high growth for a time and restricts the middle class, is an excellent policy from their point of view of the ruling class in China. So I don’t see there being an objection from the point of view of the rulers of that nation, and in fact I see a great benefit for the rulers of that nation.

Setting that aside, though, only if you believe that high growth would be unobtainable at all without the currency being kept undervalued does that policy make sense, and for a nation with the natural resources and population of China, such a claim is absurd. China, acting in a free trade environment and allowing free markets internally, would grow rapidly no matter what because of the Chinese people’s, and in fact almost all people’s, inclination to barter and truck for their own advancement.

The central planners of the Chinese economy have taken the stolen wealth and put it into projects like the Olympic complex and entire vacant cities in western China; while some may call this “growth”, I call it a misallocation of capital that will eventually lead to the bust of the Chinese economy. Additionally, instead of wasting capital on vain projects like this, if instead every Chinese worker’s bank account were worth more because the currency contained in that account was worth more, these people would each be able to allocate their capital as they please, towards investment in their own business or towards greater consumption. And price signals would guide their decisions much more effectively than the whims of the central planners of the Chinese economy. So, while “growth” as measured by econometric statistics may be lower (although as I said above I think these statistics are quite flawed because the “growth” they measure if not real), what growth there was would be better, more sustainable growth that would benefit China more in the long term because it would be made with the benefit of better information. One of the unseen costs of this policy of undervaluing the Yuan to promote higher growth will be the eventual bust of the Chinese economy and the misery that will inflict on the people of China and, most likely, the people of the entire world.

I would argue that the United States, which did not pursue such a policy during its highest growth years a century ago, benefited from the wealth that accumulated in the hands of its middle class. And, obviously, the United States began with almost no capital compared to the tremendous wealth developed in Europe over millennia, yet it was able to grow and eventually surpass Europe. If it wasn’t necessary for the United States, it certainly isn’t necessary for a nation like China to manipulate its currency to keep labor costs down and wealth out of the hands of its workers.

There are unseen costs to all of this apparent benefit, and my point is simply that we should try to understand them to gain a more complete picture of the costs and benefits of this policy of undervaluing the currency of a nation. Additionally, if you view a republican form of government graced by individual liberty like the United States had in its prime to be a consummation devoutly to be wished for all who can obtain it, than having a middle class that is powerful and can stand up for itself because it has the wealth of the nation is also to be wished for.

Or, as Hayek famously wrote: “The curious task of economics is to demonstrate to men how little they really know about what they imagine they can design.”

If my analysis is fundamentally flawed, please elaborate on why these unseen costs I’m trying to take into account are merely figments of my imagination.

Apr 14 2011 at 1:33am

Yes, interesting discussion. But I wonder how much our perceptions will change in the next few years. We hear about China’s 10% annual GDP growth, but what about the fact that personal income has been growing at only between 2% and 4%? What does all that growth consist of, and who is benefiting? Investments by the giant “State-owned Enterprises,” SOEs, has been huge during China’s post-crisis stimulus, but what happens if and when it turns out those investments are producing a low or negative return? I suspect we’ll be re-evaluating the Chinese model in a few years, just as we re-evaluated the Swedish model and the Japanese model when it turned out the rosy pictures were obscuring growing problems.

Pietro Poggi-Corradini
Apr 14 2011 at 10:44am

“The logic is this: you want to subsidize the incumbents. What I think a lot of governments do wrong is go ahead and subsidize the copycats as well. Which, given the economic logic I just laid out, you want to subsidize the initial investor, not the subsequent ones.”

This is a highly counter-intuitive statement. In essence it says, don’t worry about everything we know about the “ratchet effect”. How policies, and especially subsidies, once in place get captured by the firms that are subsidized. How even bad and disastrous policies can live forever, and in fact creep ever more invasive, even after the initial proponents have disavowed their claims. I’d call this the “fine-tuning” fallacy. It’s similar to the idea that an enlightened dictator could inject a large stimulus into the economy and then take it all out at just the right moment. In the short-run these subsidies might have a small yet non-negligible effect, but in the long-run they could end up making things worse.

Apr 15 2011 at 9:28am

Rodrik draws an analogy between investing in R&D and investing in a foreign country, but the economic logic behind the two is not be the same, because there are more unknowns in the second case than the first.

If I successfully create a new business in Jamaica, I release much less information into the public domain than if I introduce a new product in a developed country. The product can be picked apart regardless of time and place, but other investors will have a much harder time figuring out all of the good decisions I made to create that successful Jamaican venture, and even if they do, circumstances could have changed so that their attempts to copy might fail.

I don’t think investors are worried about competitors rushing in after they set up a successful business in a developed country; I think they’re worried about whether that initial investment is even going to pay off.

john berg
Apr 15 2011 at 12:06pm

@Mads Lindstrom @mike @pietro poggi-corradini

I don’t want to distract from my point that the US is an exceptional nation because of its Constitution and that endogenous forces seem to want to destroy the Constitution by mocking it. The US Patent System gives a subsidy, I suppose, in the form of a legislated monopoly. But it gives it to the inventor and only as an opportunity to pursue. The inventor get no direct money subsidy and the inventor must pursue the correct actions to produce the benefits. The fact that some wish to make our patent system like the European system and change our Constitution may result in eliminating the exceptional nation that proved a lifesaver for Europe and the world at least twice in the 20th century.

John Berg

Apr 15 2011 at 1:30pm

Great podcast.
Hearing debates about development is always interesting and always leaves me with the same familiar impression:
economist do not take seriously the political economic dynamics in certain countries, particularly African countries.

African and Latin America were sort of lumped together in this podcast, but I think African countries are strikingly different from most
developing countries at least for one reason: economic development is for the most part absent of the internal indigenous
debate. May be Asian countries got it right, and to some extent Latin American countries were on the right path at some
point, simply because the local people really care about economic developement i.e. simpy because they had a plan.

In sub-saharan african countries there is no plan. Planning for what could be the best path towards development is not even
remotely part of the political debate(when there is one), or part of the concerns in the civil society(when there is one) or anywhere.
Most every body is a rent seeker to some extent, and the medium to long term planning is completely absent. Who’s doing the thinking, IMF and World Bank.

Apr 17 2011 at 8:24pm


Your argument does nothing to substantially refute Rodrik’s claims empirically.

He is claiming to have evidential support for the fact that countries, most notably in Latin America, who have both opened trade AND floated their currencies freely, and drastically cut back on subsidies, etc. have stagnated, while those who have not, notably in Asia, have not experienced these issues.

There is only one of two ways to argue this, demonstrate that these factors were not the cause of the disparate growth rates, or demonstrate that these growth rates are misleading, or unimportant.

Rodrik’s claims seems very plausible to me as Asia’s booming economies have resulted from a combination of freeing up of trade practices, while maintaining strategic subsidies, protections, and currency policies.

I understand the basic Anarcho-Capitalist sensibility you are coming from, but you have to do more than declare the evilness of states to with the economic argument here.

Apr 17 2011 at 9:29pm


One more thing, as a “for example.”

If the best way to increase the average person’s REAL wage in a country is to devalue that currency generally, wouldn’t that be justified and desired? Or do you think out of mere principle it would be better to be poorer (able to purchase less goods and services) but with a strong dollar of which you have very few?

It seems to me people are concerned more with their absolute general wealth and welfare than the specifics of the dollars valuations as compared to other currencies. After all, if I give you 100 x $10 dollar bills in exchange for 300 X $5 dollar bills I will still take the trade even though the individual bills I am exchanging are valued for less.

Apr 18 2011 at 12:53am

@John Berg

Obama has already presented his independently verified certificate of live birth. This document can be used to get a passport and other government services as proof of native birth.

In addition Hawaiian officials have confirmed that the long form birth certificate is on file.

So why does his documented proof of Hawaiian birth not count, and do you really think we are going to have a rush of potential presidential candidates whose birth place is ambiguous?

Apr 18 2011 at 4:23pm

Well, if I asked anyone, what’s the difference between Asian Tigers and Latin America, I guess I would get many answers concerning work ethics, corruption, monetary/political/law stability…

Would protectionism/subsidies would just miracously change all this? Alternatively, was really the work ethics, corruption and the stable environment comparable between these countries?

psycholog warszawa
Apr 19 2011 at 8:51pm

I scout’s honour subsume to your masterly approach. These are pieces of damned helpful tidings that wishes be of downright foster need of in behalf of me in tomorrow

Apr 21 2011 at 1:52pm

@Andy (also my name by the way 🙂

“Would protectionism/subsidies would just miracously change all this?”

I don’t think anything “miraculous” is being proposed. He seems to be
stating something that he feels fits most inline with the empirical
evidence he’s seen. He could be wrong, but I don’t see a reason on
it’s face to describe it as in some way miraculous, unless you believe
government policies are incapable of effecting the economy, which I
think few would agree.

As to his actual prescriptions, his claim seems to be that if you
compare countries which liberalized (in the libertarian sense) their
international and domestic trade quickly and without taking into account possible strategic concerns, then these economies tend to stagnate and experience low growth. If on the other hand countries liberalize their economies in a more measured manner, keeping some of their current subsidies and control of their currency, while also eliminating many trade restrictions, that these countries according to the evidence he has seem to do better.

“Alternatively, was really the work ethics, corruption  and the stable
environment comparable between these countries?”

I think it would be generally agreed I would hope that no two or more countries represent a perfect isolated experiment. All countries have their own cultures, histories, traditions, etc. but that does not discount gaining any information from these instance I don’t think, do you?

For instance, if you wish to throw out any empirical comparison of countries one could easily claim that authoritarian communism could still work in some countries, and the problem was just the work ethics, corruption, and stable environment in which communist governments originally took hold.

I on the other hand wish to argue that we can gain knowledge, if imperfectly and continuously revised, from comparisons. In fact I am surprised there isn’t more international economics comparisons and discussions. If one has a problem with these international comparisons they should also remember that the United States culture, environment, work ethic, etc. have all changed from decade to decade WITHIN the United States so it would be fallacious to claim this to be insulated from the difficulties of cross-country comparisons.

I think it is perfectly legitimate to look at a countries policy choices, follow the outcome and do ones best to both take into account other possible effects on the outcome, as well as making a coherent and predictive story as to how the policy results in an effect. As more and more instances are observe in such a way, somewhat more confidence can be placed in the assertion, of course always subject to revision and arguments of bias, poor informational basis, etc.

Comments are closed.


About this week's guest:

About ideas and people mentioned in this podcast:Books:


Podcasts and Blogs:



Podcast Episode Highlights
0:36Intro. [Recording date: March 11, 2011 (accidentally says 2009 in the mp3 file)] Globalization and its impact. Start by quoting from recent paper you wrote with Margaret McMillan:
Clearly, globalization has facilitated technology transfer and contributed to efficiencies in production. Yet the very diverse outcomes we observe among developing countries suggest that the consequences of globalization depend on the manner in which countries integrate into the global economy. In several cases--most notably China, India, and some other Asian countries--globalization's promise has been fulfilled. High-productivity employment opportunities have expanded and structural change has contributed to overall growth. But in many other cases--in Latin America and Sub-Saharan Africa--globalization appears not to have fostered the desirable kind of structural change. Labor has moved in the wrong direction, from more productive to less productive activities, including, most notably, informality.
Talk first what you mean by structural change and its relationship to productivity. In the paper itself, the measure of structural change that I'm using is one that is based on a 9-sector disaggregation of the entire economy. So, that means I have all of agriculture in one sector; all of manufacturing as one sector; all of, say, minerals, as one sector; the government sector in one sector; and bunch of other service sectors. So, it's very broad patterns of structural change that I'm focusing on. Going back to a very early tradition in development economics, that goes back to Sir Arthur Lewis's work in the 1950s, when he developed these models of the dual economy, and he defined the process of economic development as one where basically labor moves from the more traditional parts of the economy--traditional agriculture--to more modern, more productive parts, typically urban centers, industry as a leading modern sector. I wanted to see what has happened to that idea, to what extent that process still captures what is going on in the developing world. And there were some surprises, which you mentioned in the quote that you gave. That transition from rural to urban is going on all over the world, right? It's definitely happening in massive amounts in China, in large amounts in India. Is that the general direction also in Latin America and Africa? To a large extent that is true. I think last year was the first year when the total number of people of the world in urban areas of the world exceeded the number of people in rural areas. There is generally a trend toward urbanization; that is for sure. There are countries in our sample where that is not happening, including at least one African country surprisingly. But that is in general the trend that is happening. What we capturing in general is the fact that some of the countries in our sample are able to increase employment in some of the more productive parts of the urban sectors, whether those are some of the manufacturing sector or some services closely linked with the manufacturing sector. Those are countries where the structure tends to promote overall productivity because the economy's resources are moving in the right direction. Other cases, where, if you look only at, for example, how manufacturing has done in a lot of Latin American countries, you walk away with the impression that it has been quite an extraordinary performance. In fact, that is one of the most surprising things in the paper. A lot of the work that has focused on the performance of specific industries in Latin America, for example, during the period since 1990 when the regional [?] opened to globalization, when you look at how manufacturing had fared in response to opening up to trade. There is a very common finding in this literature, which as the firms where the productivity increased tremendously, it increased more, the more subjects individual industries within manufacturing were subject to the forces of international competition. Which sort of stands to reason. But what this literature looked at: the fact that in many of these manufacturing sectors, increases in productivity takes place through technological upgrading, through increased capital accumulation, and not through employment increases. In fact, often, in fact, employment shrinks. So, we get this phenomenon of rationalization, so that part of the increase in productivity is due to the rationalization of production. And, this wouldn't be a problem if the labor that is displaced from these rapidly, productively-increased firms go to other parts of the economy. But that's really what hasn't happened. And that's what our research shows: that there has been a massive movement in the wrong direction from these more productive, more open parts of the economy to the less productive services.
6:54Just to recap for a minute: If you see in some aggregate that productivity is growing, it could be because there is more output, but what we often are going to care about is output per worker. If output per worker is growing dramatically because of investments in technology in capital, but actually the number of workers could be falling because of the productivity of capital, the question is: What do those displaced workers do? In a developed country--and I want to come back occasionally to contrast similarities in your story with the developed country--we often talk about the virtues of creative destruction. So, new technology comes along, it gets cheaper or more efficient to produce some good; and the workers go to find something else to do. Total output expands; there is more choice, more variety, things for lower prices, higher standard of living, etc. So, the question is: How well do labor markets work--part of the question--in these different economies? Where do these workers go? What do we know about that? Where do displaced workers go when these economies become more productive. First of all, let me say that I think this distinction between what seems to be happening in the developed and developing countries is very important. There has been a tremendous amount of structural change in the advanced countries as well. In particular, manufacturing employment has shrunk as a share of total employment, pretty much in all industrial countries. Now, this in itself isn't reflected in overall reduced growth in productivity for the economy, because people who are employed who are displaced from manufacturing in the rich countries don't necessarily go to sectors where their productivity is significantly lower. Could be a little lower. But, there is a chart in the paper which shows that the inter-sectoral variation in labor productivity tends to significantly come down over the course of development. So, the inter-sectoral variation in labor productivity is huge in the developed countries, in part the developing countries, in part the lowest-income among them. If you look at what happens to those inter-sectoral, the coefficient of variation of these labor productivities across different sectors; and by the time you look at the rich countries, the gaps have really shrunk. Why is it that these gaps aren't being equalized? If you turn the question on its head and say: This is the big question about development. In every developing country you have a whole bunch of activities that are actually very productive. The problem is that they are a very tiny part of the economy. The question is: Why isn't labor and other resources moving sufficiently rapidly to those activities?
9:57Give the example, the story of Malawi. The mining sector--here is a sector that is very capital-intensive, and therefore labor productivity is very high. In fact, so high in Malawi that it matches output per worker in the U.S. economy as a whole. So, in fact, if you could stuff all of Malawi's labor force into the mining sector, Malawi would be as rich as the United States is. Of course, you can't; nor would it necessarily make sense to employ all those workers because of the amount of capital it would require. But the fact is that the workers you cannot get into mining, in agriculture, formal activities, petty trade where their productivity is just a tiny fraction of what it is in mining. Let me ask a more basic question more generally; the Malawi example brings it out. If you look at many manufacturing facilities in the United States, a small number of workers are producing enormous amounts of output. The two industries I've looked at--factories I've actually toured--pencil manufacturing and the egg production of chickens. In both those industries, the number of people involved has shrunk dramatically in the United States. Some of the jobs have left the United States; but still a lot of production of chicken eggs in the United States. Hard to ship. In those facilities, a very small number of workers produce a billion eggs or more per year. There's a temptation to say they are very productive. And they are, in terms of average productivity. But it's really not the workers that are productive. It's the machinery that is smart and allows relatively unskilled workers to work in that factory and look like the source of productivity. In fact, what's really going on is that the labor embedded in the machinery is where the productivity lies. The workers are there mainly just to see that the machinery doesn't break down. They don't really do anything other than keep an eye on things. When you think about Malawi, even though the productivity might be very high relative to the United States, those workers are not relatively skilled or productive--I assume--they are working with unbelievably skilled technology and machinery that could be done by many types of people. So, their true marginal product is not very high. It depends. In principle, if these economies were working like textbook market economies--and I said earlier, the textbook puzzle is why these gaps aren't closing--the gaps that ought to be closing are gaps in the value marginal product of labor. If markets in these economies worked well, and if the governments didn't impose artificial impediments on labor moving across sectors, then what you would expect is that wages and value marginal product of labor would be equalized across different sectors. That might mean at the margin it would be beneficial for the mining sector to want to expend, hire more workers, given the huge gap that there is in wages between what they effectively pay in mining and in other sectors. I think there is a difference here with mining, precisely because it so capital intensive and it depends so much on a particular resource, it has limited employment absorption capacity. If you quadrupled your mining by hiring labor and capital, you'd lower the productivity, perhaps very dramatically and perhaps wouldn't be able to sell it at the same price. Exactly. So, that's the sense in which I think the Malawian mining example is only interesting in terms of the orders of magnitude; but not as a counterfactual for what the Malawian economy ought to look like. The more interesting question is with respect to factors they know they can expand. So, you look to an economy like Ethiopia. Surprising that many things that ought to be produced and could be produced in Ethiopia are not being produced. There are matches--being imported from Kenya. Cardboard boxes imported from other countries. There is no reason why Ethiopia cannot produce these things. And you know that little factories that produce simple consumer goods like that would hire workers and pay them multiple of what those workers can produce in agriculture. Yet there is something in the economy that you don't get these new industries, that prevents these industries from taking off or absorbing more labor. How much of that is due to what we would sometimes call red tape? Hernando de Soto, I think the World Bank, have done extensive work on it; and they find that in many small countries it's extremely difficult to start a new enterprise. Is that part of the story? Well, the big question here is, and you can think about why these gaps aren't closed: two kinds of explanations. One is government failures. Governments are screwing things up, and red tape would be a clear example of that. Others mentioned would be corruption. If you are operating petty trade, the government cannot come and tax you; you have a workshop somewhere--fix the boat and then you are subject to the government taxing you, inspectors coming and asking for bribes and so forth. You might figure it's not really worth your while. And more generally, property rights and the overall enforcement of contracts. Rule of law. If there is no rule of law, then you are okay working as a barber on the corner store because everything you need is personal relationships; you don't have to buy a lot of inputs, you don't have to deal with Customs, tax inspectors all that much. All these problems with the government and legal and contractual environment clearly impose disproportionate costs on the modern economy relative to the traditional part. So, that's one reason why the modern part is underdeveloped. So, that's sort of under the rubric of government failures.
18:11But I think often market failures are very important as well, and those are things like coordination failures: if you want to operate a cardboard factory, you need high-quality paper; you need particular kinds of workers who are used to showing up at work and doing a good job. You need all kinds of complementary inputs and arrangements around you for this to make sense. You have pineapples, but you don't have pineapple canning. So, you say: Okay, I can export this. But in order to think of exporting canned pineapples, you need a national airline or some airline to have cargo service; and they are not going to establish a cargo service until they see a demand for it. All kinds of learning externalities as well. There is a problem, what we call cost discovery: the first investor who invests in, say, the cardboard factory provides a very valuable signal to other investors as to whether the cost structure of the economy is conducive to producing cardboard or not. If you end up being a failure, all the costs are private. If you end up being successful, then you are sending a positive signal and you cannot fully internalize all the benefits of that because others come in on the basis of that valuable signal. So there are learning externalities, coordination externalities under the general rubric of market failures. The list is long, of why these gaps exist and don't close sufficiently rapidly. Obviously the trick is in specific contexts trying to figure out which are the most binding constraint in any particular setting. Sometimes it will be market failures, sometimes government failures. I can give you examples of both. A few years ago we were in El Salvador, an economy which has done a tremendous amount in terms of liberalizing its economy, privatizing, stabilizing. Nothing essentially wrong that you can point to in terms of the contractual legal environment. It wasn't the government that was just hell-bent on taxing everything. Quite the opposite. And nothing except for garments had taken off; and garments had taken off because of some special trade privilege in the U.S. market, but nothing except for that. You ask people: If I give you $25 million, what would you invest in in the Salvadorian economy, and they would think about it a little bit and then they would say: Well, can I put it in Miami? Do I have to invest in El Salvador? That's the kind of economy where, when you don't see any Schumpeterian rents--we forget. When we think about Schumpeterian rents, we think it's just about rich countries. Rich countries: Schumpeterian rents are important for innovators, because they get a new product and they need to ensure they have at least some profits from investing in those high risk activities. But in developing countries too there are unique Schumpeterian rents for investors willing to go into new areas, new industries. Capital likes to get a return. Exactly. And when it's risky and when there are huge spillovers to the rest of the economy, the private return is going to be way below what the social return is, so you are not going to get the right kind of transformation. In that kind of a setting it's not government failure that's blocking transformation. It's just a bunch of market failures associated with low levels of income. And there you do want the government to come in and actually do some stimulating.
22:32Let me ask you about that, because I'm a little skeptical of that terminology, calling that a market failure. It's very uncertain. You could say the same thing in the United States. You gave me $25 million; I wouldn't put it in Miami, by the way. I'd think where's the best place to put it, and of course it's not clear. If you make me put one bet down, the equivalent statement to putting it in Miami would be I'll put it in an index mutual fund. That's also not much of an answer. If I had to put a bet down on a particular sector, that's hard to know. That uncertainty is inherent in any economic situation. But I agree that it's qualitatively different--and obviously quantitatively different--in a poor country. In a poor country the decision is hard for a different reason. In a rich country, everything's pretty similar. You still could make a mistake, but the expectation is pretty similar. There, it could be an enormous return that's waiting if I just knew which leap to take, which direction. Is it cardboard boxes? Is it something else? It's certainly not clear the government could do it any better. Part of it is just the uncertainty, but my story goes beyond the uncertainty because as you say, if it's just the uncertainty there's no market failure. The market failure arises from the fact that there is no way that the pioneer investor in a new area can recoup the full social return that it generates. The parallel you want to think of is this: In a rich country, we have a patent system that says if you come up with a new product, we are going to allow you to earn monopoly rents from this product. Why? Because if the knowledge you generate becomes freely available to all then you earn no rents on this product; and then you lose all the investment in R&D that you made, and therefore the competitive level of profits isn't enough to pay you for the investment in R&D. So, we need to give you some rents. The patent system is meant to overcome this market imperfection in the market for knowledge. The parallel in developing countries is that when we are talking about firms who are not going to be investing in new products, but in new products that are new to their own economy; and that provides the same kind of learning spillovers as a new product in a rich country does. So, the first investor that sets up a call center, let's say in Jamaica, and points to Jamaica as a place where you can successfully run call centers for the universe of all potential firms, that initial pioneer investor has provided a huge amount of social surplus to Jamaica. I think it does make sense for Jamaica to want to subsidize the initial investor in that. And often those investments will not pay. It's just the nature of this. All based on the idea that most of these investments will actually not work out. So, some of these will be mistakes. But the economic logic of this is you want to subsidize the incumbent investors in new area. I disagree. I'm going to make one more attempt; you can counter. The reason I disagree is there's a marginal and an infra-marginal decision there. One is to open a call center at all; the second is how big to make it. Either of those is uncertain. You don't know whether a call center is going to pay off, and you don't know if it should be big or small, how much to invest. But it's very easily the case that if the return is very high--there's uncertainty, we agree; that part you can't get rid of. It's true you won't capture all the return. There's a bigger return beyond that to all the investors because they've learned that Jamaica has a skilled labor force. And you are taking a chance on that; you are giving evidence that that's the case. Or, you are giving evidence about the reliability of the Jamaican government to keep it's promises. There are a lot of different aspects and margins these decisions could be relevant for. Could provide information beyond just your narrow amount. But if you make an enormous amount, the fact that there's spillovers doesn't really mean that you need a subsidy to do it. You just don't capture all of it. It's true you might not make as big a decision, the right size decision; but even if that's true, even if I'm wrong about this, I still don't see how a government subsidy to that call center is going to be a good idea, given that the government is probably going to use political motivations in setting that subsidy. There's no reason to think they are going to make the right decision. You made two points: one is that it didn't make sense economically; the other is that even if it did, it's politically not going to work out. I think economically the argument is this: You are right that the extent that the investor subsequently earns some degree of extra-normal profit, then he's willing to engage in the search cost on behalf of others who don't necessarily have to incur that search cost. But think of the call center. That's a competitive industry. So the initial investor who is going to try to figure out whether Jamaicans are really suitable for providing that service of a call center is incurring an investment, and if he is successful, if subsequent investors can come in and bid his workers away--and this being a competitive industry, he can only look forward to normal profits subsequently--then he has no incentive of coming in. So, if there is enough extra profits for some reason, which happens for another market imperfection--if the markets are insufficiently competitive--so he can rely on markets not being sufficiently competitive to provide those incentives. Those developed countries are getting into the kinds of areas where they are competing with everybody. Margins are small, so you can't rely on these investments--the guy who is going to be providing canned pineapples to the U.S. market or call centers or the guys producing cut flowers, all new industries in these kinds of settings, necessarily to earn supernormal profits once they come in. Yet from the perspective of the economy they generate huge social surplus because once you demonstrate that cut flowers can be profitably grown in, say, Ethiopia, you have an area of employment where productivity is much larger than it was elsewhere. So, I think there is a good economic reason, if the problem that's blocking the rise of new industries is what I've called these cost discovery externalities, that you want to subsidize. The logic is this: you want to subsidize the incumbents. What I think a lot of governments do wrong is go ahead and subsidize the copycats as well. Which, given the economic logic I just laid out, you want to subsidize the initial investor, not the subsequent ones.
30:47But this is only one of the economic arguments why you may want to use certain kinds of industrial policy. The second argument about whether you can do it or not, whether political motives get in or not, I'm happy to talk about that. Let's move to something related and you can bring some of those ideas in. I think the most striking point in the paper, and I think the central issue facing the debate over globalization today is we have these two incredible success stories of China and India. Some smaller but still dramatic successes, most of them in Asia. What do we understand about what they are doing right and other countries are doing wrong, or could it perhaps be structural, internal to the nature of the country? Some of it is policy-related, obviously. Do we know anything about why it's so different? I'll put forward a hypothesis, I think consistent with the story in the paper, a story that requires more empirical substantiation. But it's a story I find in line with what I see. You need to compare the way in which China opened itself up to the world economy and globalized with the way that Latin America did. Latin America did it in a way that was consistent with our standard trade theory. Standard trade theory says you lower your tariff barriers, your income-competing firms that are not sufficiently productive shrink, and the labor that's released from those industries go to a whole bunch of other places--notably, your export-oriented industries, where you have a comparative advantage, where their productivity at the margin is going to be higher. So, in fact, you get from opening up to the world economy an overall higher level of labor productivity in the economy because you are reducing this misallocation. Now, if this story doesn't work quite in this way because of all the problems we've discussed just now--there are certain obstacles, whether they are market failures or government failures that prevent the sufficiently rapid rise of new industries where you are going to be really competitive, then what happens is what we observe in Latin America, which is that the import-competing industries, true, they have to become more competitive, so they do shrink; but the labor that is released doesn't necessarily end up in industries that are sufficiently more productive. That's exactly what we observe in the data; what we have called this growth-reducing structural change. How China differs from this pattern is that they have opened up very differently. They have opened up at the margin. So, look at the structure of China in the 1980s when they began to open up. They had a whole bunch of state enterprises operating inefficiently, behind barriers to trade and huge support from the state. The standard strategy for liberalization would have been to tell them to just remove those barriers, remove those supports. Stand on your own two feet. And then you will create these new industries. That's not how they've opened up. They kept all those industries in place, all those supports in place, but they created at the margin incentives for new investments in export-oriented areas. So, they created special economic zones; they created all kinds of subsidies for non-traditional products in exports at the margin. They turned toward world markets without pulling the rug from underneath these industries with medium level of productivity, when you look at the economy as a whole. These state enterprises aren't as productive as firms in the special economic zones, but they are still much more productive than the rural areas, or doing petty trade. So, that's the problem they have avoided, which Latin America has not. So, then, I guess the question would be--I'm not an expert on China--is how much of China's employment growth in urban areas, which has been extraordinary: we're talking hundreds of millions of people in the last 15-20 years moving from rural activity to urban activity--how much of that is new enterprises either started domestically or by foreign investment? That's the thing that is not happening with the displaced workers in Latin America or in Africa. Do we have any idea? As opposed to the expansion of existing ones that are already kind of doing stuff? Right. Now, I can't tell you in terms of numbers of enterprises, but it is clear that what has happened, for example, that the share of employment in manufacturing has steadily increased in China way beyond--only recently begun to shrink. Whereas in Latin America, for example, it began to shrink prematurely. So if you look at cross-country benchmarks in terms of what happens to employment and manufacturing, over the course of development, what you experienced in countries like Latin America is a premature de-industrialization. In China, if anything the other way around. So, what I can tell is that China managed to provide incentives for the growth of its manufacturing enterprises. We know there is a huge amount of new entry, but I can't give you a specific number on that from a comparative perspective. But, as you point out, interesting way to think about it. We can think about sectors having different productivities within a country, but we can also think across countries. One of the things that's happened in the last 15 years is the opportunity to profitably produce things in China is sucking capital and technology into China, away from Brazil, African countries that might have produced those things. Plus, there are some externalities of shipping and infrastructure that China has been able to create or has had before. As a result, productivity increases there are driving the flow of capital there, and now we've only gotten to the point where those are being exhausted and we are seeing a decrease in Chinese manufacturing employment as a percentage of the total because as technology continues to improve, you don't need as many people to make the stuff.
38:46But that's very depressing. For every poor nation struggling to deal with this, it used to be a manufacturing center--Mexico is having a tough time. They are close to the United States, which is a great advantage, but not as cheap as China. So, they've got to offset that price difference with the transportation savings, and they can't always do that. It still raises the question: given that the world is an increasingly dynamic place, you want to make it as easy as you can for innovative, creative people, whether they are your own citizens or others, to find ways to employ people, start enterprises. Absolutely. The problem is that many of these enterprises in countries in Africa or Latin America are being successful aren't generating a lot of employment, so this is precisely the problem. There is another side of this, which is the world of the exchange rate. I say that there is no better industrial policy than an undervalued currency. This is one thing that Latin American countries have systematically gotten wrong, including most notably Mexico. The one thing that has repeatedly happened in countries that have let their currencies float and basically had monetary regimes that say we don't care where the exchange rate goes, just a market-determined variable; and you open up your capital account, money comes in; your currency appreciates. And it basically kills new investments at the margin in tradable industries, manufactures. That's what's happened throughout Latin America. Asian countries are very careful to prevent that from happening. Of course, maybe China has gone overboard with undervaluing its currency. In terms of just this business of promoting new industries, I think part of it is industrial policy, and we can discuss how the effect of that is. But part of it is a competitive currency. Industrial policy has the problem that it becomes politicized very quickly, and you have to figure out who you are going to select. The beauty of the exchange rate, of devaluation, is that it's across the board. You don't have to pick. If your currency is undervalued by 10%, all tradable industries are being subsidized across the board by 10%. I think that helps a lot. Furthermore, when your currency is overvalued, because of capital inflows and so forth, no amount of industrial policy can really undo the effect of that. A 20% overvaluation of the exchange rate--there is no way you are going to undo it or compensate for it by microeconomic policy. Microeconomic policy can get easily swamped by these macro prices. That again has been a very big difference between Asia and Latin America. So, what happens in Latin America is then they get locked into a pattern of specialization that is much more intense in natural resources and primary products; and those sectors don't generate a whole lot of employment. You also generate a lot of political rent seeking. Bad for governance, bad for politics, for that reason. It's a lousy dynamic. I don't think it was all that preordained that China or other countries in the region, and Taiwan before, would have necessarily ended up being manufacturing powerhouses. They decided to become manufacturing powerhouses, and they became successful at it. I think it's somewhat misleading. I think what's really happened is the technology is driving that. It's not that China got good at making things. They allowed people to bring in the capital that made their citizens the vehicle for that change. A friend of mine involved in the retail clothing business said 30 years ago, maybe even 20 years ago, a sweater factory in China was a bunch of women with knitting needles. There's a limit to how productive you can be with a pair of knitting needles. You can go a little faster. But the room for improvement is relatively small. That's not how they make sweaters in China any more. They make them with the same machinery I suspect they would use in South Carolina, to the extent we still have some textile factories in South Carolina. I don't think we still have any sweater factories.
43:46Let me ask you one question about the currency thing. I'm agnostic, skeptical about it. In the United States, people claim that Chinese currency underevaluations hurt manufacturing jobs; and China's strategic use of currency is helping them and hurting us. Strange argument for us. We can maybe talk about whether it's different for Ethiopia. But for the United States, if they are subsidizing their currency and making their stuff cheap for us to buy, we are getting a big improvement in the price level that allows us to have a larger accumulation of material stuff. That's an increase in our standard of living. True, it displaces workers potentially in manufacturing; but our labor market does work pretty well. They get put into other things, and basically that is going to be good for the United States on average. I can understand how in Mexico it might not be, if their labor market doesn't work very well. Seems like the right answer to that is to try to improve their labor market and not get into a currency war. I think you are largely right about the United States. In normal circumstances, given that labor markets work well, there aren't huge discrepancies at labor productivities at the margin across different types of economic activities--you might say the structural production in the United States is relatively immaterial; it doesn't really matter. If China wants to be less huge in terms of trade gains--if they want to give us clothes, toys, seems good. Moreover, since labor markets there don't work so well, and there are huge gaps between productivity in the manufacturing and rural sector, with the reserve army of surplus workers in China--for them, structural production does matter. From a global standpoint, it's a very efficiency-enhancing policy. Now that makes perfect sense, up until the point where you get in the United States a 9% unemployment rate, and then you have to start asking the question to what extent that is still true. If you believe there is some element of Keynesian unemployment in the United States today then the argument about the external deficit of the United States being a gift to U.S. consumers is only partway true. Keynes became a protectionist in the 1930s. Something I find mysterious and strange. But it makes sense. Without retaliation. Well, there is a coordination problem of course if everybody does that. So, the argument here would be it's far better to reach an agreement with China on the exchange rate than have this kind of protectionist war. You are right also that even under normal circumstances this is a bigger issue for developing countries like Mexico because for them, too, the structure of production matters. You might say, well, let them deal with better labor market policies. But I think one thing I've learned working on developing countries is you are always in a second-best world. Often we just say: Pick the first best. That just becomes often escapism, because you are inheriting often a second-best world. Given that that is not the only distortion you are talking about, everything is entangled with something else; and often the reforms that work best are those that are second-best reforms that tend to isolate some adverse situation that would have happened somewhere else in the economy. You have to be more pragmatic there. I don't know. When it's that complicated, I'm not sure what pragmatic means.
48:27Let's go to the United States. You said a few minutes ago something I've started to hear people say about the United States, which I find a little strange, but it certainly is parallel: which is that the sectors that are growing in the United States--recently that has been the technology sector--are not very labor-intensive. Google and Facebook, the most dramatic successes in the technology area, hire tens of thousands of people, but they don't hire much more than that. So the people ask where is the new employment going to come from? How much do you think our anemic labor market as part of the recovery of the recession is structural, versus other possible explanations? Just that we put a whole bunch of people into housing? We pulled a whole bunch of people into housing between 1995 and 2005. That was easy. The world went long kind of well. Then all of a sudden the economy found itself with too many carpenters, electricians, drywall applicators, and they had to find something else to do; and they seem to be struggling to do it. What's going on? There is definitely that structural misallocation given what the U.S. economy has inherited, so there is definitely a need for a structural adjustment going forward; and definitely growth is going to be slow until you get that readjustment. I do not study the U.S. economy, so I don't have a very good sense about this debate. I find it hard to believe that there is not an element of Keynesian demand side unemployment when unemployment shoots up to that level. I am inclined to believe there is a Keynesian element in this. But it's more an impression and a guess than based on substantive analysis. I appreciate the honesty. Not sure anybody knows very much about it. We certainly hear a lot more dramatic claims along either side. I want to introduce a parallel. One of the issues that has come up on this program before is that in general when people dump goods in the United States, free trade economists say: That's good. But somehow when we dump in foreign countries that's not good. Why is it that the logic isn't the same? What we've been talking about today is that when labor markets don't work very well, whether it's government or market issues, some of the standard conclusions of trade policy may not be similar in poor and rich countries. The general question I want to ask you is about globalization. Currently in the current climate we are in, which is not so healthy economic activity--high unemployment, in lots of countries. Where do you think the global economy is headed? You are renowned as a skeptic and contrarian on trade issues. Try to give us a summary of what you think is important in these areas for both policy and what you think is happening. I think we've come out of the crisis with a sense that you need much better regulation, for example of financial markets; that financial globalization has gone beyond the regulatory infrastructure that you needed to sustain it, and that therefore you have to complete that regulatory infrastructure. So we have efforts, G-20, the Basel Committee, Financial Stability Board, the IMF, and so forth. If you just use the analogy with domestic financial markets, we've learned we just need a whole alphabet soup of regulatory authorities that we are still fine-tuning to try to get it straight. Behind all of that there is a whole political structure. Expand that logic to the global level, and there is no way we are going to get anything like the global equivalent of domestic regulations that we have to sustain domestic finance. Given that, either we are going to get regulations at the global level that will be very weak, ineffective, until the next crisis; or we will get the lowest common denominator because that's the only one you can get countries to agree to. So I think that's the path we are moving down. Ensuring that we minimize transaction costs associated with national borders and global finance; that we need to coordinate and harmonize at the global level; and I think that's a dead end. If you agree with me, then the only alternative is to have a global financial system that allows a much greater diversity and discretion at the national level in terms of how these tools are set. Some countries may set their capital requirements at 20%, others at 10% because they want to be on a different point on their financial innovation versus financial stability frontier. I think we should allow that. The problem then becomes how do you deal with the regulatory arbitrage that would arise from different countries having different kinds of regulatory regimes. And there I say: We should approach this issue in exactly the same way we approach our imports of toys from China. When we import toys from China, we say they better the bite[?] by their own health and safety requirements--if they do not conform to lead content requirements then those goods don't come in. Finished. There is not even a debate about that. I don't think there is any difference with financial markets. Our rules should be the same: that any financial institution that wants to operate in the U.S. jurisdiction ought to operate by the rules set in the United States. And, in order to maintain those rules, the integrity of U.S. rules, you ought to then allow the United States and other countries to interfere or tax or otherwise control cross-border financial flows to ensure that their financial systems maintain integrity and isn't eroded. So it's a somewhat very different conception of where we ought to go. But I think it's really frankly the only realistic one. So, we shouldn't allow toxic toys; we shouldn't allow toxic assets.
56:18Let me partly agree with you and then partly disagree. I agree with you that the harmonization thing is a bad idea. For a different reason. Maybe we agree on this. We actually have had quite an extraordinary global financial infrastructure. Basel I and Basel II were the products of thousands, maybe millions of hours' work by experts. You could argue that maybe next time we'll get it right. But I would argue maybe it's not designed to be gotten right. It's a process that is inherently going to be run by the people with the closest, biggest stakes, the most information. Those would be the biggest international financial firms. Why would you expect them to give inputs to the political process that would be good for the world rather than good for them? If anything, going to a globally harmonized system, and this is true for accounting and many things: The claim is made that it's going to save all these transactions costs. It could. But there's no reason to think that that would be the way the system would be designed. I assume it would be designed by politically powerful people to benefit themselves. Given that there is no accountability, who would we blame for Basel II, that didn't work so well? Who do we get mad at? Who loses their job? Nobody. So, there's no reason to think that Basel III, even with a bunch of well-intentioned people, would do better. Or Basel IV. Whatever we are at. I agree with you 100% on this. I think this is the problem we've had with the system. Economists tend to think that by moving these decisions from the national to these global fora, you have depoliticized the decision-making. That's a good point. But what you've done is just privileged one set of people over everybody else. That's exactly how the system has worked; it is exactly how the system is not working with Basel III. That's why I think you should politicize the system! It should be at the national level where the full force of politics can play out. That's again an argument that is pushing in that direction. How about goods and people, though? You answered my question with respect to financial flows. There it's very clear. We've eked out most of the benefits from further liberalization. We are wasting a lot of political capital at huge cost in terms of legitimacy and so forth, and I think some quite legitimate concerns about the distributional concerns of what are suffered. And people, they are sort of like, where we were in the trade regime in the 1950s, which is a huge amount of gain to be had. So, I want to sort of see all those trade negotiators in Geneva stop talking about the current [?] and start talking about the current temporary work resource schemes around the world, where the gains are huge.