Sunday, September 18, 2011

Great Stagnation...or Great Relocation?


"I was on top of the Westin Hotel being shown the sights of [Shanghai], and I had a sudden crisis as I looked out at the extraordinary skyscrapers the architecture and the art deco. I thought to myself, well, the mandate of heaven has passed from us and come home." - Gore Vidal, 2007

I'll say this about Tyler Cowen's Great Stagnation hypothesis: it has really made me think. Although I was resistant to it at first, the more I look at the evidence, the more compelling it seems. Income growth in the U.S. really has slowed down since the 1960s, and more recently, in all the rich countries.

Cowen's idea has much to recommend it. It's undeniable that the easy gains from universal public education and exploitation of cheap land are over. And it also seems pretty clear that there has been a stagnation in transportation technology since the mid-century, driven by the plateauing efficiency of our energy sources. There's a reason our innovation has switched from stuff like cars and planes to stuff like computers and phones.

On the other hand, the idea that our stagnation is driven by exogenous changes in scientific discovery - we just didn't find enough new stuff this half-century! - does not sit easily with me. One reason for this is that I have a very hard time believing that changes in the rate of technological progress could cause a decrease in per-capita incomes. Consider this graph of real working-age household income since 1987 (courtesy of Karl Smith):


I just have a hard time believing that a slowdown in scientific discovery could cause incomes to go down. I mean, technology doesn't get worse over time, right? (This is also one of the big problems that everyone has with RBC models.) I mean, mayyyyybe technology could stagnate so much that we hit a Malthusian ceiling, in which population growth increases resource costs to the point where people start getting poorer. But it just seems unlikely.

Therefore, I have been thinking about alternative theories to explain rich-world income stagnation. And I have come up with something. I call it the "Great Relocation". The idea, in a nutshell, is that economic activity is relocating from rich Europe, America, and Asia to developing Asia faster than technological progress can replenish it.

The idea of how this works is not due to me, but to Paul Krugman.

Economic Agglomeration and the New Economic Geography

In 2008, Paul Krugman won the Nobel Prize, in part for a theory that called the New Economic Geography. That also happens to be the first economic theory I ever learned in detail (back in 2004!), and the one that lured me into the field of economics. It's a brilliant idea, and has  some solid empirical support. You should read about it, if you don't mind a little math.

The basic idea of the theory is this: It is expensive to move products around. This means that if you have a factory, you want to locate it close to where your customers are, to avoid paying a bunch of shipping costs. Now consider two factories. The workers in the first factory will be the consumers for the second factory, and vice versa. So the two factories want to locate near each other ("agglomeration"). As for the workers/consumers, they want to go where the jobs are, so they move near the factories. Result: a city. The world becomes divided into an industrial "Core" and a much poorer agricultural "Periphery" that produces food, energy, and minerals for the Core. 

Now when you have different countries, the situation gets more interesting. Capital can flow relatively easily across borders (i.e. you can put your factory anywhere you like), but labor cannot. If you start with a world where everyone's a farmer, agglomeration starts in one country, but that country gets maxed out when the costs of density (high land prices) start to cancel out the effect of agglomeration. As transport costs fall and the economy grows, the industrial Core spreads from country to country. Often this spread is quite abrupt, resulting in successive "growth miracles" that get faster and faster (as each new industrial region starts out with a bigger global customer base). The evidence strongly indicates that agglomeration is the driver behind developing-world growth.

But here's the thing: in the theory, the "old Core" doesn't keep getting richer. In fact, under some scenarios (which are difficult to explain concisely), the old Core even gets slightly poorer while the "new Core" catches up. For a while, the negative effects of relocation trump the positive effects of progress.

So this could explain why people in the rich world are getting poorer. In the 50s, America was the only industrial "Core" on the planet. But since the 60s, we have seen successive "growth miracles": Japan and Europe in the 60s/70s, then Taiwan/Korea/Singapore in the 80s, then China since then, and now even India. In a New Economic Geography world, we would expect these successive relocations of manufacturing to hold down income growth in the U.S., even if technology was advancing as usual. And now Japan and Europe are feeling the pinch as well.

Core and Periphery

But the problem is especially acute for America, and here's why. Let's think about Cores and Peripheries. Take a look at this population density map of the world:


Suppose all of those people had the same purchasing power. If you were a factory owner, and you wanted to minimize transport costs, where would you put your factories? The answer is a no-brainer: China and India. Some others in Europe, Japan, and Indonesia. Perhaps a couple on the U.S. East Coast. But for the most part, you'd laugh in the face of any consultant who told you to put a factory in the U.S. The place looks like one giant farm!

It may be that American manufacturing strength was due to a historical accident. Here is the story I'm thinking of. First, in the late 19th and early 20th centuries, our proximity to Europe - at that time the only agglomerated Core in the world - allowed us to serve as a low-cost manufacturing base. Then, after World War 2, the U.S. was the only rich capitalist economy not in ruins, so we became the new Core. But as Europe and Japan recovered, our lack of population density made our manufacturing dominance short-lived.

Now, with China finally free of its communist constraints, economic activity is reverting to where it ought to be. More and more, you hear about companies relocating to China not for the cheap labor, but because of the huge domestic market. This is exactly the New Economic Geography in action. 

Sustain Points and the Great Recession

Actually, the story could be even worse for us. There is another interesting feature of the New Economic Geography theory: hysteresis. I.e., history matters. A place that becomes a city will not easily turn back into a farm. There is a "sustain point" in the economic forces that make an agglomeration viable, and as long as a Core region stays above that point, its initial good luck in becoming the Core will keep it safe from turning back into a backwater.

BUT, a severe shock can knock you down below the sustain point. If your Core no longer has a good reason to be the Core - if it remained rich and built-up only because it would have cost too much to move it - then a big recession (or a devastating war, or a natural disaster) will cause economic activity to leave permanently. Think about the permanent shrinkage of New Orleans in the wake of Katrina.

In this light, the Great Recession of 2008-whenever might be a lot more ominous than even the Great Depression. What if this gargantuan shock has finally made it worthwhile for the center of world industrial activity to relocate to the North China Plain, where nature says it ought to reside? 

In that worst-case scenario, the U.S. industrial economy is not coming back any time soon. From the air we look like a farm, and that is what we will once again become. Thanks to our technological edge, we will retain strengths in industries like software and business services, for which transport costs are extremely low. But our days of "good jobs for everybody" are done. (Keep in mind, this is just a worst-case hypothetical.)

Relocation and Stagnation

Note that although this Great Relocation is an alternative to Tyler Cowen's Great Stagnation, it does not preclude it. Lower productvity growth could coexist alongside agglomeration effects. Or...they might even go together. As I wrote in an earlier post, some "endogenous growth" theories suggest that the availability of cheap labor can reduce the incentives for innovation. If technological progress has stalled, it might just be because the Great Relocation has taken priority.

So What Do We Do About It?

So China "took our jobs." But this was not due to their exchange rate policy, or their export subsidies, or their willingness to pollute their rivers and abuse their workers, although all these things probably spend the transition. They took our jobs because it made no sense for a farm like the U.S. to be building the world's cars and fridges in the first place. Forcing China to revalue the yuan might slow the Great Relocation a little, but has zero hope of stopping it.

So how do we fight this force of nature? How do we get new jobs? I have a few ideas, but keep in mind that these are the thoughts of Smith, not Krugman, and thus take them with a couple extra grains of salt.

1. Keep immigration going strong. 

The United States still has a massive technological and institutional edge. That means that people who come here will be more productive than Chinese people for a long time to come. Every worker we add increases local demand by a lot, and decreases the incentive for production to relocate. This goes double (or triple, or quadruple) for high-skilled immigrants. Beefing up our population with the world's elite knowledge workers is pretty much a no-brainer. Canada gets this; we do not

And in the long term - a few decades down the line - high immigration levels will even reverse the Great Relocation, if China's population ages dramatically and ours remains young and growing. With our fertility at exactly the replacement level, the U.S. will only grow by immigration.

2. Promote urban density.

Agglomeration happens because people live close to one another. But the United States has some of the world's least-dense cities, thanks to massive zoning and building height restrictions (and free parking). If darker purple on the "density map" equals long-term prosperity, we need to scrap these policies and start encouraging high-density housing and efficient light rail systems. We need a place to put all those new, high-skilled immigrants!

This will require conservatives to give up their nonsensical anti-train animus. It will also require certain well-off liberals to drop their NIMBY-ish insistence on "open space" (I'm looking at you, Silicon Valley!). Most importantly, because new immigrants will come mostly from Asia, this will require Americans of all stripes to reconcile themselves permanently to the notion that they live in a multi-racial country. 

3. Repair the roads, now.

Agglomeration depends not only on density but on transportation costs. Lower transportation costs between cities make it easier to transport goods between cities. If we make it cheaper for California and Michigan to supply each other, it strengthens the status of the United States as a single industrial agglomeration.

Our roads here in America used to be the world's best, but now they are falling apart for lack of maintenance. This needs to change, and now - while borrowing costs are at rock bottom - is exactly the time to do it. Note that public goods spending is also believed to be the most effective kind of stimulus...so this is a case in which there is zero conflict between short-term and long-term priorities.

4. Conclude free trade deals with other rich countries.

The Great Relocation story is about the world's industrial Core shifting to poor countries in Asia. This may cause some to agitate for trade protectionism (though I think this could only slow the transition, and probably at great cost). But it makes lots of sense for us to boost our trade with other rich countries - i.e., the other countries of the Old Core. Production is not going to relocate from the U.S. to Japan or Europe, because costs there are equally high. Instead, what will happen is closer to the standard "comparative advantage" or "New Trade" stories, where both countries get a boost. (Another way of thinking about this is that free trade lets us improve our competitive position vis-a-vis China when selling stuff to Europe and rich Asia.)

There is one such agreement currently on the table: the Korea-US Free Trade Agreement. Korea is a rich, developed country, so we should pass this right away.

(Update: I removed a fifth suggestion, "support basic research," since I am kind of a broken record on that one anyway, and it's not really agglomeration-related.)

So those are my policy suggestions for fighting a Great Relocation. Keep in mind that these things are almost certainly good things to do no matter what. But in a world where the U.S. is in danger of turning back into a farm, these measures become essential.

Also, note that the Great Relocation is a story about the long term, not the short term. These are not policies to fight the current recession; they are policies to increase the growth of U.S. median income over the next 20 to 50 years. 

Anyway, there is my alternative to the Great Stagnation story. My story is equally scary, but it is something that policy can address. I am not certain that this is really what is going on, but I think it it is worth a lot of thought, and some research too.

OK, back to dissertation. :)


Update: Ryan Avent writes:
It's not clear what story Mr Smith is telling.
Let me boil it down. There are two stories here:

Basic story: Rich countries are experiencing a temporary and mild drop in living standards as China and other poor countries go through an ultra-rapid transition from farming to industrialization.

Scarier story: With the entry of East and South Asia into the global trading system, the U.S. will be in the "agglomeration shadow" of the new pattern, and hence will (partially) deindustrialize. This transition was hastened by the shock of the Great Recession.

I'm not saying either story is definitely true; I don't have the empirical work to make a definitive statement. The basic story sounds very plausible to me, and the scarier story sounds less likely but very scary. My point is agglomeration is a big idea and nobody so far has been talking about it much when they talk about developed-world income stagnation.


Update 2: The Economist has some data that seem to support this general story I'm telling. Thanks to commenter "notyourbusiness" for the link.

45 comments:

  1. I suspect there is truth to Tyler Cowen's ideas, although I suspect I am biased toward believing that because, being in the software business, I've long expected we would create this problem sooner or later. What bugs me about it is best explained by asking, "So why is Tyler Cowen a libertarian?"

    If I believed the problems he identified were the primary issues preventing a recovery of jobs, then some policy implications seem obvious to me: 1. If job shortages are here to stay, then we should forget about raising the retirement age and, if anything, try to get baby boomers to retire sooner so more millenials have opportunities to leave their moms' basements. 2. If the private sector can grow its productivity and meet new demand without creating jobs, then stimulus measures should focus on growing the public sector rather than trying to convince the private sector to create jobs it doesn't need to create.

    Needless to say Tyler Cowen doesn't see it that way. I think his take goes more like "The problems we are facing have a root cause that we aren't going to solve, therefore we should treat addressing these problems as futile and learn to accept a world of high unemployment and low middle class wages." Sort of like how it is futile to rebuild Japan since we really can't do anything to stop earthquakes.

    But why is anyone a libertarian? It seems to me there are two reasons. 1. A belief in free market efficiency. But never mind whether libertarianism actually maximizes economic productivity. If Tyler Cowen is right, if you believe that the economy is now so productive it doesn't need a large chunk of our involvement, why do you even care about making it more productive? 2. A belief that libertarianism = freedom. But I don't see why anyone would be so attached to a system that has no role for a large chunk of the population to play and as a result does not provide for them. Is that really the ideal of liberty?

    Now a lot of this may be irrelevent if Tyler Cowen is wrong, but I'd be curious to hear from anyone who thinks he's right -- what is the point of libertarianism?

    ReplyDelete
  2. Nice post. I always like to see Fujita-Krugman-Venables discussed, because it is one of very few genuine advances in macroeconomic theory since Keynes's death.

    This topic - stagnation in per capita income - is being debated in New Zealand economics circles, although probably the less mainstream parts so far.

    In brief, the New Zealand Paradox is that despite world-leading or top-ten scores for institutions (freedom from corruption, transparency, property right protections, ease of doing business, education of the work force, work force flexibility, labour force participation, working hours, etc., etc.), New Zealand's per capita GDP is not growing at the same pace as the rest of the OECD's. In fact, NZ is on pace to drop out of the OECD sometime in the 2030s.

    The economic geography school (e.g. Philip McCann, now at Groningen) have an argument like yours, Noah, and also note that scale matters -- there seem to be increasing returns to scale. The roles of trust and information spillovers are also noted. The main policy prescription is to make NZ appear to China to be both closer and bigger than it is, and to more tightly integrate with Australia to this end.

    One sub-school advocates increasing urban density as well; another stresses Richard Florida's arguments about the mobility of the creative, entrepreneurial class. On this argument it's essential to create places that high-tech entrepreneurs want to live, rather than just increasing density.

    The 'cultural factors' school is post-NEG. It says that transport and communication costs are now (or will be soon) so low that geography is becoming less important.

    This school holds that the cause of the paradox is cultural: New Zealanders are choosing to work in low-productivity service industries rather than in high-tech, high-margin manufacturing. (The NZ prime minister's advocacy of NZ becoming a "banking back office" for Asia-Pacific is a case in point. He seems unaware that per capita output is feeble in service back offices. Which is why they are moving to low-cost places like India and Egypt.)

    This idea, that the causes of stagnation are partly cultural, is an interesting one and worth further exploration.

    However, I don't think either of these ideas apply very strongly to the USA, which is by itself one-fifth of world GDP. Surely this is big enough to constitute a core of its own.

    That brings me to the third school. This is an 'economic factors' school which holds that the cause of stagnation is malinvestment. Taxation in NZ has privileged real estate over every other investment, and capital gains over income. The idea is that investments that return income create greater employment than does speculation on capital gain.

    On this view the appropriate policy is to tax capital ownership, to re-balance investment decisions in the direction of cash flow versus capital gain. This would appear to be relevant to the US.

    Anyway, thanks for this post.

    PS. FWIW, I can't see any evidence of a technological stagnation.

    ReplyDelete
  3. An alternate view... the Great Stagnation happens to coincide with the period where we abandoned even the pretense of Keynesian economics. A few graphs here:

    http://www.angrybearblog.com/2011/02/another-look-at-keynesianism-and-great.html

    ReplyDelete
  4. Greg: Interesting stuff. Has NZ considered large-scale immigration? I'd say that's the key, coupled with an export-oriented industrial policy focused on goods with low shipping costs, like software, business services, etc. With the right social engineering (see Austin, Texas for a model), NZ's govt. could make it a VERY attractive place for the intellectual elite of America, Japan, and Europe to move (and of course immigrants from China and India are in infinite availability and have good skills as well!).

    If NZ wants to hire me as an economic advisor, I could work wonders... ;)

    ReplyDelete
  5. New Zealand is very pro-immigration, most skilled people wouldn't have a problem getting in.

    It has also attempted to increase density in its cities by limiting expansion, making places like Auckland largely unaffordable to the average New Zealander.

    I wonder if they are trying too much of a boutique strategy, looking too much to Manhattan and Silicon Valley as models instead of Houston or Austin.

    ReplyDelete
  6. This is a great post Noah. As a high skill foreign worker on his last few very thin strands of patience with the idiotic US immigration system I can personally attest to (1) how relatively unfriendly and hard to navigate the system is (2) how much this is discouraging talented people from trying find work here. The greatest irony is that folks from the new "cores" still find the US good for one thing: its higher education. Then, after graduation, they pack up and go to China and India, perhaps after a short stint at a Wall Street bank to hone their skills.

    ReplyDelete
  7. This is a fascinating theory, and thanks for posting it. Regarding your proposed solutions, I have two thoughts.

    On immigration, it bears elaborating that Canada (my homeland) takes in proportionally more immigrants than the US, but they are 1-2 in the world, and the biggest difference is that we pick almost all our immigrants, while the immigrant population to the US has a substantial element of, um, self-selecting for desperation. So you've made a pretty good argument that the US should maintain or increase current total immigration, while doing everything it can to shut down the flow of undocumented migrants.

    The second thing is I'm not sure "promote urban density" is precise, so much as "increase the US population." The largest US cities are already as large as the largest Chinese cities; Houston is populous, not very dense, and growing comfortably (even the property values are reasonable). The US model is weird by global standards, but if the US is going to grow substantially in population (and that effectively means urban population), it's not clear to me whether that growth is most easily added by making the Bay Area denser, Houston bigger, or by increasing the population of Salt Lake City dramatically.

    ReplyDelete
  8. Basic science does have a place when it comes to agglomeration (clustering) of bioscience industries, as this 2006 global map shows: http://www.nature.com/embor/journal/v7/n2/fig_tab/7400633_f1.html
    Such agglomeration has proceeded apace in the years since, particularly in China and India.

    ReplyDelete
  9. Anonymous4:28 PM

    Ryan,

    Why do you think that "doing everything it can to shut down the flow of undocumented migrants" is in the US' best interests? Obviously getting highly-skilled/educated immigrants would help the US but if the goal is to increase population then why oppose illegal immigration?

    If unskilled immigrants come to the US they are still adding to demand and having kids who will be the workers and consumers of the future. A poor person in Southern Mexico adds little or nothing to US GDP, but if that poor person moves to the US and has kids who become Americans (with American productivity levels and education etc) then the US is better off.

    ReplyDelete
  10. Anonymous7:50 PM

    I think the map of population density needs to be weighted with "purchasing power density," among other things. That seems a significant omission.

    ReplyDelete
  11. Anonymous8:21 PM

    It is expensive to move products around.

    ...in the late 19th and early 20th centuries, our proximity to Europe...allowed us to serve as a low-cost manufacturing base.


    These two statements do not compute. In what universe was the US in proximity to Europe 100 years ago?

    ReplyDelete
  12. Noah: NZ has had policies such as you describe for quite some time.

    Without high levels of immigration NZ would be in a worse state than it is, as many NZers emigrate, or work overseas for decades at a time. There are about a million NZ nationals living outside the country; the domestic population is just over 4 million.

    Most NZers welcome immigration, as it supports property prices. But they're not too keen on rapid population growth, so there's some political see-sawing on this.

    The government has long held to a no-industrial-policy doctrine, failing to recognise that the taxation regime itself constitutes an industrial policy. However there are signs that this no-policy policy is weakening, with a few targeted interventions recently.

    In the last decade the government has provided venture capital to startups -- as private savers' savings are all tied up in real estate for tax reasons, NZ has a desperate shortage of liquid capital. Mostly these startups have been in software and niche high-margin products. Like all politically controlled programs, however, the 'incubator' programme is subject to considerable uncertainties.

    If you wanted to go into the NZ Paradox, I'm sure you'd be welcomed. The people at Motu (motu.org.nz) or The New Zealand Institute (nzinstitute.org) might be good first contacts - they'll know other people with insights, too.

    But don't you have a dissertation to finish, first? ;-)

    ReplyDelete
  13. Noah- I like the post, but a couple of things don't hold up. As noted above, Krugman's agglomoration story, as rellated here, doesn't explain the US rise. Nor does it really explain China's rise to this point. They currently have the lowest rate of consumption as a percent of GDP in economic history. Agglomoration isn't primarily about being close to customers, Biogen, Biomed, or Acme Bio Lab isn't in Cambridge to be near sick people; it's about pooling specilized talent.

    ReplyDelete
  14. Greg: Why yes, I do. ;)

    OGT: Oh yes, I think agglomeration does explain the U.S.'s rise. And industry clustering is more about being close to suppliers and downstream business customers than being close to retail customers.

    ReplyDelete
  15. Did anyone actually read the Noah Smith post? Is their really any reason take it seriously? The author is suggesting that Asia is booming because higher population densities yield lower transportation costs.

    Really? Not low wages. Not a low cost, high skilled labor force. Not developed nation levels of productivity (in some sectors) combined with high productivity. Not education rising / before real wages. But lower transportation costs?

    What ever happened to the “death of distance”? The containerization revolution? The Internet? Bulk carriers? Unit trains? America is losing out in world trade because our transport costs are too high?

    In real life, transportation costs have steadily fallen (even allowing for higher oil prices) enabling “remote” production centers to competitively produce goods for rich country markets. This obvious fact essentially blows up the entire “success via population density” thesis.

    If you doubt this, check one paragraph from the Noah post.

    “It may be that American manufacturing strength was due to a historical accident. Here is the story I’m thinking of. First, in the late 19th and early 20th centuries, our proximity to Europe – at that time the only agglomerated Core in the world – allowed us to serve as a low-cost manufacturing base. ”

    Wow. That’s exactly wrong. In the 19th and early 20th centuries, American industries thrived behind high tariff barriers. Our exports were commodities. Only later did America become a competitive producer of industrial exports.

    The truth is that this appeals to NS because it provides a rationale for Open Borders, not because it makes any sense.

    ReplyDelete
  16. It was the growth of government, from 29.9 percent of GDP in the 1960s, to 32.8 percent of GDP in the 1970s, 34.7 percent of GDP in the 1980s, 35.3 percent of GDP in the 1990s, and 38.9 percent of GDP today, that has led to the reduction of economic growth rates and wage stagnation.

    If Western nations reduced the share of the economy that is centrally planned, economic growth rates would improve.

    ReplyDelete
  17. This is what is called a "falsifiable" thesis. In other words it can tested and disproved.

    If the NS model was correct, manufacturing would be steadily migrating to the high density coasts (particularly the East coast) and away for lower density states.

    The reverse is true.

    If the NS model was correct, Europe with its higher density would have higher manufacturing productivity than the U.S.

    The reverse is true. See http://economics.sas.upenn.edu/~dkrueger/research/kk-paper.pdf

    As it turns out... Low density America has the highest manufacturing productivity in the world. This should mean a global migration of factories to the U.S.

    Not happening obviously.

    Hence we can conclude that the NS thesis has been falsified.

    ReplyDelete
  18. Anonymous said...

    Immigrants do not "add to demand". Contrary to popular Open Borders mythology they don't defecate gold coins that they promptly spend boosting the local economy. Actually the reverse is closer to the truth. The compete in the job market and then send some of their wages home.

    Under the best of circumstances, immigrants increase both the supply of labor and the demand for labor. Other than reducing wages for natives via higher taxes, crime, inferior schools, congestion effects, unaffordable housing, etc. they are economically neutral.

    Of course, that is not true in the US today. For decades, immigrants have (massively) displayed natives in the labor force, making them a huge net negative.

    ReplyDelete
  19. It seems like Skeptical Economist has a bit of an axe to grind with people of color who come to the US to work...

    @Noah: this reminds me a bit of Edward Glaeser's work - is this part of your dissertation by any chance? This post seems like it could become a paper

    ReplyDelete
  20. boing3887,

    My comments were obviously race neutral. However, your introduction of race strongly suggests Open Borders is driven in part by racial special interest groups.

    Nothing new about that. Let me quote from Samuel Gompers, the founder of the AFL.

    "America must not be overwhelmed.

    "Every effort to enact immigration legislation must expect to meet a number of hostile forces and, in particular, two hostile forces of considerable strength.

    "One of these is composed of corporation employers who desire to employ physical strength (broad backs) at the lowest possible wage and who prefer a rapidly revolving labor supply at low wages to a regular supply of American wage earners at fair wages.

    "The other is composed of racial roups in the United States who oppose all restrictive legislation because they want the doors left open for an influx of their countrymen regardless of the menace to the people of their adopted country.'

    ReplyDelete
  21. Noah- OK, I don't get it. How does agglomoration explain why the US frose as a manufacturing powerhouse prior to Germany? Germany was closer to the core, more dense, and cheaper in terms of labor.

    Or this in the sense discussed here; this post about button manufacturing in China. Population and density are just not that deterministic once one gets beyond a rural/urban dichotomy.

    http://krugman.blogs.nytimes.com/2009/11/04/increasing-returns-in-a-comparative-advantage-world/

    I still actually like your prescriptions, and do not think you should have cleared the fifth point. If anything the logic of absolute advantage should make historical path more important, making research even more valuable.

    ReplyDelete
  22. Noah asserts:

    "It is expensive to move products around."

    That's becoming less and less true. Shipping costs are a tiny percentage of total costs these days.

    Ever heard of containerization?

    Can you give examples from the last generation in America of how growing urban density facilitated the growth of local manufacturing? The only examples I can think of are from sweatshops in downtown LA in the 1980s, and those are lousy jobs. How has growing population density made California, say, more of an industrial powerhouse? More population density leads to more NIMBYism because there are more backyards.

    ReplyDelete
  23. Anonymous9:24 AM

    Would just like to say that this discussion often misses the most important realities we are facing at this moment on this planet,ie; that the priorities of wealth and power and the structure that has supported them are in truth no longer sustainable. Our Mother Earth is showing us daily that we must change our priorities or perish as a species, and relatively soon. In my opinion, continuing to focus on wealth creation rather than survival by means of sustainable energy and agricultural production is a complete wast of very precious time. Wake up and smell the mass exstinctions! PLEASE!

    ReplyDelete
  24. Adding on to Anonymous' comment.

    Noah: "...in the late 19th and early 20th centuries, our proximity to Europe...allowed us to serve as a low-cost manufacturing base."

    As has been pointed out, this is the direct opposite of what happened.

    IIRC, US wages were higher than wages in Europe. Think of how many people came from Europe from the Civil War to the 1920's, precisely because they could make a better living here.

    ReplyDelete
  25. A lot to respond to here. Some general important points:

    1. In the New Economic Geography model, falling transport costs are what drive the spread of agglomeration. Containerized shipping is what allowed China to start off as a low-cost production base for the U.S. and Europe. But when that gave Chinese consumers a critical level of purchasing power, the Chinese domestic market started becoming important and agglomeration took off there. See Yukon Huang (who, I might add, has the best name in all of economics) for
    more on this transition.

    2. Transport costs include proximity to suppliers and customers. It is a lot easier to deal with people face to face when you have to deal with them regularly. Also, knowledge spillovers are much stronger when companies in the same industry are clustered close together (e.g. Silicon Valley).

    3. My thesis does not say trade barriers are bad. In fact, I think we should absolutely use trade barriers to force China to revalue its currency.

    4. Certain parties here seem unlikely to support my plan for one million Asian immigrants per year...but really, you should reconsider! :)

    ReplyDelete
  26. Noah: "Certain parties here seem unlikely to support my plan for one million Asian immigrants per year...but really, you should reconsider! :)"

    Noah, when you spend 30-odd years squeezing the middle and working classes, they tend to get worried, and jealous.

    ReplyDelete
  27. Anonymous1:39 PM

    How wll resource limitations on population influence the outcome? There seems to be an underlying assumption that growth can continue indefinitely.

    ReplyDelete
  28. @Barry: You talk as if letting in a bunch of immigrants is going to hurt the middle class. Maybe that seems obvious to you?

    See, I think that letting in a bunch of immigrants, especially high-skilled immigrants, is going to really help the middle class. Those immigrants start businesses that employ the middle class. Their labor also creates wealth that they spend here, which boosts the whole economy. And if agglomeration effects are important (as I conjecture in this post), those immigrants help keep America as the center of global economic activity, which has huge benefits for our middle class.

    Think about it.

    ReplyDelete
  29. Any number of poster have done a more than sufficient job of destroying the transportation economics logic for higher population density. Clearly transportation costs have fallen dramatically over time. Even when they were much higher, Japan was still able to emerge as an industrial power (per-WWII) in spite of Japan's isolation from the Atlantic economy of the period.

    The more serious argument is that agglomeration effects are real, even if they aren't literally based on transportation. NS mentions face-to-face meetings.

    More broadly, economic clusters are real and for good reasons. Pittsburgh was a classic industrial cluster in the 19th century. A few decades later, the south end of Lake Michigan played the same roll.

    In our own times, New York city is an economic cluster in finance (to America's regret). Hollywood is a classic cluster to anyone who has ever worked there. Obviously, Silicon Valley is a cluster.

    The question then is does America lack a sufficient population for clusters? Is 310 million not enough. Would adding 1 million more people make any meaningful difference in any plausible time frame.

    Yes and no are the obvious answers.

    ReplyDelete
  30. Make that

    "Is 310 million people enough"

    ReplyDelete
  31. NS,

    The "entrepreneurial immigrant" myth has been debunked. Take a look at

    "Reconsidering Immigrant Entrepreneurship: An Examination of Self-Employment Among Natives and the Foreign-Born"

    "Between 1960 and 1997, the self-employment rate of immigrants fell from 13.8 percent to 11.3 percent, while the self-employment rate for natives increased from 9.6 to 11.8 percent. The difference is now statistically insignificant; by 1997, therefore, the presence of immigrants had no effect on the overall level of entrepreneurship in the U.S."

    Note that some Asian immigrants are slightly more entrepreneurial than natives (Chinese, South Asians). Others are less entrepreneurial (Philippines, Vietnam). Only Koreans are highly entrepreneurial compared to natives.

    Of course, immigration myths are commonplace. Take a look at the Hispanic family hype versus the realities on the ground.

    ReplyDelete
  32. NS,

    "Their labor also creates wealth that they spend here, which boosts the whole economy"

    Under the best of circumstances, immigrants increase both the supply of labor and the demand for labor (raising GDP). Other than reducing wages for natives via higher taxes, crime, inferior schools, congestion effects, unaffordable housing, etc. they are economically neutral.

    Of course, that is not true in the US today. For decades, immigrants have (massively) displayed natives in the labor force, making them a huge net negative

    ReplyDelete
  33. For fun, I looked up that stats on a few European countries.

    Greece - 11.3 million people. 221 people per square mile. Lots of coastline. Cheap water born transportation. Direct land connections to the rest of Europe.

    Sweden - 9.4 million people. 53 people per square mile. Lots of coastline. Cheap water born transportation. Until recently, no direct land connections to Europe.

    Clearly, Sweden should be poor and Greece should be rich.

    ReplyDelete
  34. Anonymous4:56 PM

    Posted this on the Economist site, fwiw:

    If you translate the summary "economic success now depends on loosening immigration rules, making it easy to build in cities, in part by investing in the infrastructure that supports them, and continuing to support research and education" into:

    - opening up data access
    - publishing APIs to make it easy for anyone to build on the platform
    - investing in the infrastructure
    - supporting 'open' efforts

    this is exactly what many high-tech companies have done. For example, look at Amazon's AWS, Facebook and Salesforce.com

    Just saying...

    ReplyDelete
  35. Over at http://www.economist.com/blogs/freeexchange/2011/09/geography you can find this post discussed in some detail. Dissected might be a better description. Read it all. Useful quote

    "It's not clear what story Mr Smith is telling. On the one hand, he argues that transportation costs are high enough that it's worth it for manufacturers to move to Asia to be near its massive, growing markets. If this is the case, however, then America has little to fear. The North American market is home to over 500m people, many of them very rich. If transport costs are so high that North American manufacturers can't affordably serve Asia, then they're also sufficiently high that Asia can't serve North America, and there will continue to be room for a large North American industrial sector to serve the domestic market.

    If, on the other hand, transport costs are low enough that a single manufacturing hub can produce for a global market, as seems to be the case, then it's unlikely that firms are relocating to Asia just to be near its markets. Instead, they're likely attracted by low labour costs, lax regulation, and generous government incentives. In that case, America's lower population is not a reason to fear total deindustrialisation, particularly since these Chinese advantages are likely fleeting."

    ReplyDelete
  36. Skeptical Economist - Lots of good points, but I am inherently skeptical of anyone who hides behind a mysterious screen name.


    Maybe this is Noah attempting to play devil's advocate...? ;)

    ReplyDelete
  37. Noah said...

    " @Barry: You talk as if letting in a bunch of immigrants is going to hurt the middle class. Maybe that seems obvious to you?"

    Noah, I expect better of you.
    What I'm pointing out is that the middle and working class has been squeezed and squeezed hard. They're scared, and paranoid about anything which looks threatening.

    You are not excused from understanding this, even though your econ training is geared to making you not understand it.

    ReplyDelete
  38. @Barry:

    OK, now I see what you mean. Yes, I agree. Irrational fear that any policy change represents a threat. Sounds about right.

    But just out of curiosity, have you ever tried reading your comments out loud before you post them? I used to write stuff like "I expect better of you" and "You are not excused from understanding this"...then I realized I was a troll. ;-)

    ReplyDelete
  39. Anonymous5:41 PM

    suppose your story is true, why is it obvious this is a bad thing? maybe your policies will slow down growth in asia?

    ReplyDelete
  40. Anonymous3:18 PM

    Hey Noah, great post. I had not seen it from that perspective. This just appeared in The Economist:

    http://www.economist.com/node/21528979

    Some similarity to the points you make.

    ReplyDelete
  41. Anonymous12:35 AM

    I'm always curious about immigration as an "obvious" solution. In my misspent youth, I got a phd (and then attempted to find work in) a scientific field. Science is basically a worldwide endeavor, immigration is very easy, and yet science is not a miraculous field, brimming with jobs.

    The capital to employ scientists has simply not kept pace with the production of scientists, and so lots of phd physicists, chemists and mathematicians will never have a career in their field. Even those that do will face incredibly career uncertainty, etc. I work in finance, and have spent a few hours each week for years now applying to scientific jobs, hoping against all reason that someone will hire me to do the job I spent a decade training for.

    So why should we expect opening borders to somehow help solve a job problem? The standard assumption seems to be that immigrants spend money, more demand = more jobs, etc. But this sequence seems to have largely broken down right now. We aren't hurting for labor, the supply is glutted. Increasing the glut is not likely to solve any short term problem.

    ReplyDelete
  42. Tracy Lightcap5:04 PM

    Noah:

    I'm surprised you didn't mention Veblen. His Theory of Business Enterprise is quite prescient on the whole process of what he calls "the penalty of taking the lead".

    And, of course, there's Wallerstein. Sorry to suggest more reading, but there it is.

    Btw, I've read your blog from the first with interest and profit.

    ReplyDelete
  43. Anonymous10:31 PM

    A quick heads up regarding a typo: "although all these things probably spend the transition"

    Should be "sped the transition," I believe.

    ReplyDelete
  44. You forget the internet.
    It could have devastating effect on your theory.
    Even that at the moment there is not too big percentage of ppl working remote.
    In 10-20 years after the current crisis passes, remote working will be no-brainier.
    Except in purely physical labor where you have to be in the factory or the field to work.
    So agglomeration will happen on totally different plane.
    In a world where oil is depleted where the transport and renting cost continue to rise ... remote work is the winner.

    ReplyDelete
  45. If Krugman believes this, why is he a Progressive? Because one of the responses to structurally fleeing jobs and increasingly efficient productivity, is to lower any artificial cost of production that would hasten 'off-shoring.'

    Licensing laws, education costs, zoning laws, and any EPA laws that arguably jump the shark, work place liability risk, all should fall under careful scrutiny. IOW, does the regulation add value or hamper it? We can take negative externalities like pollution into account and still ask the question.

    The West has artificially and significantly raised the cost of living and the cost of doing business in the last couple decades, even as it pretended that we can retire for 25 years fueled by a tax pyramid scheme instead of investment. The good news, for example, is that our air is 70-80% cleaner than it was in the 1970s.

    The bad news is that even industrial land must now be as pristine as a forest to sell it to another industrial user; hundreds of thousands of acres lay abandoned because of that standard.

    One of the implications of Krugman's argument is that the world in general will get richer together; brittle economies will hasten their own decline.

    And to promote open borders in a welfare state is clearly arguable. This is no longer the land of 2 acres a mule, no matter how much we long for those days, or pretend they are still here. Also, we allow predominantly unskilled labor into the country, jack the divorce rate, and then bemoan living standard stagnation. Huh?

    Economies are complex, and pretending government regulation itself has no negative externality is a kind of blindness. Culture matters. So does artificially induced standards of living and increasing regulation which eventually returns no marginal value.

    Part of the issue of macroeconomics begins with its attempt to make sweeping generalizations from bad measurements at shallow depths. It would be like trying to characterize or manage a company by looking at total sales. It can not be done.

    ReplyDelete