Here is my interview on The Bob Zadek Show. Thanks for having me on Bob. I’d appreciate any comments on how I can improve my interviewee skills.
Yesterday I spoke with Tyler Cowen about my offer to bet that free cities would lead to growth miracles. So, here are the terms.
I bet Tyler Cowen $5 that within 40 years a free city, defined as a city which imports or creates a legal system distinct from the host country, will average 10% per capita GDP growth over 20 years.
Here are some of my other predictions. I expect the first free city to be announced within 3 years. I expect the first free city to reach 100,000 residents within 12 years. I expect there to be at least 5 free cities with populations over 100,000 in 20 years. Each of these predictions I hold with around 60% certainty.
Cowen also has an excellent article in Bloomberg View. He argues that “Nations can be Startups.”
Israel, Taiwan, Hong Kong, Cayman Islands, Estonia, and South Korea could also be thought of as startup nations. Most emerged from war, civil war, the evolution or dissolution of a prior imperial or colonial relationship, or some combination of those factors. In each case there was a chance to start anew and to have founders impose a distinct vision on a new political unit. Just as we doubt that Bill Gates could have founded and grown Microsoft within the confines of the older IBM, so did the success of Estonia require freedom from the Soviet Union and Russia.
Cowen does mention special economic zones and political decentralization, though he focuses his argument on new sovereign nations. I would reverse the focus. Startup political units are likely to consist primarily of semi-autonomous cities, with only a few new nations.
The rise of China is the greatest humanitarian success story since the World War II. Over 700 million people were lifted out of poverty. Much of China’s success is due to special economic zones (SEZs). SEZs allowed China to improve their legal system on a local level, without threatening the political equilibrium.
China is exporting their SEZ model. In 2006 the Chinese government indicated its plans to expand the SEZ model worldwide. They wanted to create 50 SEZs. Deborah Brautigam and Tang Xiaoyang published a paper in 2011 on China’s SEZs in Africa, where Chinese SEZs abroad are proceeding most rapidly. This video by the Economist gives a more recent look.
There are currently six open Chinese SEZs in Africa. They are located in Egypt, Ethiopia, Mauritius, Nigeria (two), and Zambia. They focus on a variety of industries, from copper processing in Zambia to textiles in Egypt. Some are completely owned by Chinese companies while others are joint ventures. Chinese companies lead the projects, with help in negotiations by Chinese embassies. Some host governments or companies have ownership shares, but Chinese companies remain the main actors.
I was unable to find much discussion of what the SEZs actually entail. Given the lack of discussion as well as the connection to China I suspect they are traditional SEZs. This means they likely have a one of more of the following; tax incentives, simplified business registration, and expedited customs.
Recent reports suggest the SEZs are not having their expected impact. The zone in Mauritius, for example, remains empty today, a decade after it was announced. Initial problems involved a group of farmers living in the zone who refused a buyout offer. Basic infrastructure for the zone was completed around 2011, but was insufficient to attract companies.
Another example of an Chinese companies investing in SEZs abroad is in Georgia. A Chinese real estate developer, Hualing, refurbished an old Soviet factory and was granted a Free Industrial Zone (FTZ). FTZs in Georgia benefit from several forms of tax exemption. Unlike Africa, this Hualing has been successful in developing their FTZ. That being said, it remains unclear whether the FTZ is part of Chinese strategy, or if a Chinese company merely took advantage of pre-existing FTZ legislation.
Hualing is also building a new city in Georgia. Currently 1,000 families live in the new city. It does not appear to have any SEZ status.
Broadly speaking China’s exportation of their SEZ model is positive, but insufficient. The SEZs have had limited success, certainly not comparable to the success of SEZs in China.
Free cities, legally autonomous cities, can improve on the missteps of the SEZs. Free cities import successful legal systems. Charter Cities are the best known version, but free cities can be privately administered as well.
Free cities offer two potential solutions which SEZs are unable to address. First, free cities can offer a home to the world’s rapidly urbanizing population. Second, free cities can improve governance in the developing world.
The UN estimates there will be an additional 2.5 billion urban residents by 2050. While many of these residents will live in existing cities. Many will turn small villages and rural lands into the cities of tomorrow. Shenzhen, for example, grew from a fishing village of 30,000 people to have 16 million residents in the greater metropolitan area.
The developing world is plagued by governance issues. SEZs tend to improve governance on the margin. However, having a blank slate to import best governance practices would lead to more investment, more job creation, and faster growth. For example, a new city in sub-Saharan Africa which uses common law would have one of the best legal systems on the continent, becoming a destination for the youthful and the ambitious.
SEZs improve things at the margin. However, the relative failure of SEZs in Africa, both Chinese and otherwise, shows bolder thinking is needed. Free cities can rapidly improve governance while providing residential areas to the new urban population. Free cities are a viable next step in the evolution of SEZs.
Tyler Cowen recently argued there are no more growth miracles. He writes,
The East Asian growth model, for all its wonders, belongs to history. Slow and steady may be the only option left. For whatever reasons, few countries have been able to scale up their educational successes as rapidly as the East Asian tigers. Trade growth, which exceeded overall output growth in the late 20th century, now seems stagnant. Many export industries are automated and hence don’t create as many middle-class jobs as they used to.
Cowen is mistaken. Most of the growth miracles, with the possible exception of China, were based on institutions. Singapore and Hong Kong, for example, grew rapidly. Both city states used common law and as a result escaped poverty.
To argue there are no more growth miracles implies no country or region will rapidly improve their institutions. On a national level this might be the case. Countries with the capacity to rapidly improve their institutions could have done so. However, countries can contract out institutions on the local level.
Charter Cities are an example of this. By taking land where no one lives and importing successful institutions, rapid economic growth can be achieved. More likely than Charter Cities, however, are proprietary cities, autonomous, privately governed cities.
There are over 100 new city projects worldwide. These new city projects are, slowly, realizing the importance of institutions to attracting businesses and residents. Barring an unexpected event, this trend will continue. As such, the new city projects will continue to learn the importance of institutions, gaining greater autonomy, and increasing economic growth.
I would be happy to bet Tyler that over the next thirty years at least one new city will use common law and achieve economic growth over 5% per annum for a ten-year period.
Omar Al-Ubaydli wrote an article arguing the Middle East should embrace Charter Cities. He sees Paul Romer’s appointment as Chief Economist of the World Bank as an opportunity to pursue more radical reforms than otherwise available. Omar is right about the need for the Middle East to pursue autonomous cities, but they are likely to be administered privately, rather than by a developed country.
The Middle East has the largest concentration of new city projects. Dubai was the first and most successful. More recent examples include Masdar and King Abdullah Economic City. The new city projects tend to be built privately, like King Abdullah Economic City, or through public private partnerships.
Unfortunately, most of the new city projects took the wrong lessons from Dubai. They saw Dubai’s magnificent architecture and thought, “to build a new Dubai we just need to build amazing buildings.” What they fail to realize is that Dubai’s buildings did no cause its success, but are rather a consequence of it. Dubai became successful by creating special economic zones to attract investment. Their financial center, for example, runs under common law.
The future of free cities in the Middle East most likely lies with proprietary cities, private autonomous cities. The new city projects will learn, hopefully not too slowly, the importance of institutions in development. Privately financed projects are unlikely to abdicate administrative authority to a developed country. Rather they can simply hire a judge and adopt the institutions, retaining control while importing the best aspects of institutions of developed countries.
Wired has an article on moonshots, schemes that can potentially have big impacts on humanity. Unfortunately, Wired’s vision is constrained. The moonshots listed are technological feats, build an engine that can get to mars, create AI, reduce malaria. By focusing on technology, Wired misses how changing the rules of human interaction with free cities can have monumental effects.
Below are the criteria listed for the moonshot.
In February 2016, Teller laid out the principles of the philosophy. A moon- shot, he said, should be firstly about solving “a huge problem in the world that affects many millions of people.” That could cover solving the energy crisis, cutting road deaths or halting global warming. Second, a moonshot should not settle for half-baked measures: it has to provide a “radical solution” that can do away with the problem for good. The last criterion, Teller explained, is the reasonable expectation that technology can actually solve the problem. Moonshots should be as much about pragmatism as they are about dreaming.
Free cities fit this description. First, urbanization and poor governance cause poverty which affects billions of people. Free cities can alleviate that poverty. If special economic zones (SEZs) in China accounted for only 10%, a conservative estimate, of China’s economic growth, then SEZs are responsible for lifting 70 million people out of poverty.
Second, technology is the driver of long run economic growth. If free cities can even slightly accelerate technological innovation, they can improve the lives of billions of people.
Third, free cities are both radical and practical. They are radical in the sense that such devolution of power is difficult to achieve political. They are practical as free cities such as Hong Kong and Singapore have already shown it is possible to lift millions of people out of poverty.
Free cities are a moonshot, but a moonshot worth investing considerable time and energy in. I hope Wired considers free cities projects the next time they write about moonshots.
This past weekend I attended the EAGlobal, the annual effective altruism conference. I came largely to promote the idea of free cities, as I think free cities are one of the best poverty alleviation tools.
First, some sociological notes. Effective altruists, broadly speaking, fit into three camps; existential risk, animal welfare, and anti-poverty. Existential risk seems to be primarily concerned with AI. Animal welfare with factory farming. And anti-poverty with malarial nets and deworming. While there is overlap between the groups, most people I talked to tended to strongly identify with one cause.
What surprised me was the lack of interest, or even awareness in economic growth. Economic growth is the best anti-poverty tool the world has ever seen. Anti-poverty discussions in EAGlobal tended to focus on tools that had level effects, not growth effects. Deworming and malarial nets are great, but long term economic growth is better. I suspect the lack of focus on economic growth is the difficulty in measuring the impact of various ‘interventions’ on economic growth as well as the more ideological nature of the inquiry.
I got the impression there was more discussion of policy this year than in previous years. Unfortunately, much of the policy discussions involved speakers from government or think tanks, standard mainstream stuff. For example, one speaker said they were looking for solutions to eliminate the kidney shortage. Of course, there is an easy but politically unacceptable solution, legalize kidney sales. Given the willingness of effective altruists to tackle problems involving existential risk, I was disappointed to see such unimaginative policy discussions.
On the positive side, there did seem to be an interest in free cities. Most people I talked to seemed aware of the idea, or at least similar ideas like Seasteading. There was less hesitancy in discussing the idea than in a Brookings event, for example. Several of the people in leadership roles demonstrated knowledge in free cities, which makes me wonder why no speakers were invited to opine on the topic.
In sum, I think the lack of awareness of the impact of economic growth on poverty is the primary barrier to effective altruists showing more interest in free cities. Until economic growth is adopted as a primary motivator of effective altruists, I doubt they will embrace free cities.
As I’m attending the effective altruism conference currently I thought I’d share an essay I wrote, originally published on Medium, on why effective altruists should support free cities.
Effective altruism has done a great deal of good. The use of the scientific method in evaluating charities has clarified the primary causes of human suffering, while the research into existential threats illuminates the most urgent threats to civilization, if not humanity. However, effective altruists have missed the middle ground between alleviating immediate suffering and defending humanity against catastrophes, economic growth.
In this essay I argue that thinking about the causes and consequences of economic growth is a valuable use of time for effective altruists. More specifically, I argue that effective altruists should advocate for free cities as a mechanism to improve governance and promote economic growth. A free city is a city with substantial institutional autonomy from the host country, for example Hong Kong.
It is difficult to overstate the importance of sustained economic growth. Humanity, for most of its history, has lived in poverty. Images which are now associated with rural sub-Saharan Africa were the norm for most of our ancestors. It was only recently, the industrial revolution, when humanity began to escape the Malthusian trap.
Economic growth did not instantly translate into economic development, but it did slowly improve the lives of those experiencing it. The wealth created by economic growth then translated into vaccines, universal literacy, safe childbirth (for both the mother and the infant), clean water, electricity, and virtually all the benefits people currently take for granted.
Economic growth is also responsible for the greatest humanitarian story since the World War II, the rise of China. The World Bank estimates that Chinese growth lifted 500 million people out of poverty. This accounts for a far greater reduction in suffering than all Western aid efforts.
Not only is economic growth responsible for the standard of living people have today, economists have a fairly good idea about the necessary conditions to for growth, private property rights, rule of law, and the provision of public goods. Countries with private property rights, the rule of law, and that provide public goods can rapidly grow, escaping poverty in a generation, and becoming part of the developed world within two. Singapore and Hong Kong, for example, grew from impoverished islands with per capita GDP of less than $500 in 1960 to become world class cities today.
The rise of East Asian tigers illustrates the difference between growth and level effects. Eliminating malaria, for example, would save millions of lives and greatly reduce suffering, a worthy goal. However, eliminating malaria would not lead to catch up growth. Per capita GDP might double, or even triple, but sub-Saharan Africa would still remain poor by modern standards. Economic growth, on the other hand, would close the income gap, as well as ultimately helping to reduce malaria and other dangerous diseases. Europe eliminated their diseases, smallpox and polio, after they got rich, not before.
Unfortunately, few, if any, effective altruists are even discussing economic growth. Instead the focus is on one time impacts reducing human suffering, malarial nets for example, and reducing the probability of existential threats to humanity. These are both worthy causes, but should not encompass the entirety of effective altruism efforts.
An old joke about economists also explains why effective altruism chose these causes. An economist, Smith, loses his key in a dark alley. Adam, his friend, finds him there, searching under the one street light. Adam asks Smith, “did you lose your key here?” Smith responds, “no.” Adam then asks, “why are searching here, if you lost your key elsewhere in the alley?” Smith responds, “this is where the light is.”
The joke illustrates that economists, and I believe effective altruists as well, tend to focus on problems that are easy to measure. Calculating disability-adjusted life years for one time interventions is relatively simple. If malarial prevention works, there are X fewer dead children and Y fewer sick children. This can be compared to a different charity, building houses for the homeless in America. Getting the homeless off the street might save Z lives a year. If $100,000 saves 500 children from malaria, but only 5 homeless people, malarial prevention is more cost effective at reducing human suffering.
Existential threats are similarly easy to quantify. Ignoring future populations, an existential threat would kill 7 billion people. Even if the probability of the existential threat is .1, reducing the probability to 0 saves the equivalent of 7 million lives.
Jump-starting economic growth, on the other hand, is far more difficult to quantify. What are the benefits of an anti-corruption campaign? The short term effects, more efficient use of government resources, can be quantified. But what about the long term effects. What if the anti-corruption campaign increases trust in government, which in turn increases the investment in public goods, which opens markets and spurs economic growth. The causal chain is difficult enough establish, let alone quantify.
Similarly, imagine a country reduces barriers to entry for new businesses. The new businesses will increase productivity. However, the established businesses which previously had a monopoly might react by undermining the government. The outcome is difficult enough to rationalize ex post, let alone predict. The dynamics of most countries governance are sufficiently complex attempts to quantify them are close to meaningless.
This, of course, does not mean no attention should be paid to issues of governance. However, it is important to recognize that not all problems are explicable through randomized control trials or simple thought experiments. If anything, the added complexity of issues surrounding governance implies more attention should be paid to it, not less.
With that in mind, I would like to argue that free cities are one of the best mechanisms for improving governance, and by implication, jumpstarting economic growth. Most economists would accept the previously made claims, they are relatively uncontroversial. The following arguments, on the other hand, are not widely accepted. Nevertheless, I believe the creation of free cities is one of the most important, and underappreciated, ways of reducing poverty and human suffering.
A free city is a city with substantial institutional autonomy from the host country. Institutional autonomy includes independent commercial courts and an independent regulatory system. The closest modern example is Hong Kong, which is ruled under a different system than mainland China. Free cities are a way to rapidly improve governance in parts of poorly governed countries.
As previously mentioned, Hong Kong and Singapore both experienced rapid economic growth over the last 50 years. Shenzhen is a similar example, growing from a fishing village of 30,000 people to a metropolitan area of over 18 million. Its growth was largely due to its initial designation as a special economic zone, legalizing foreign direct investment in China for the first time in the modern era. Thirty years ago Dubai was a desert. Now it is a booming metropolis because it created a governance system which encourages economic investment and growth.
However, improving governance is difficult. One reason is governance quality exists for a reason. For example, attempting to reduce the power of a monopoly will cause the monopolist to retaliate. Another reason is asking a countries government to improve itself is like asking coal to turn itself into a diamond. A country with no history of good governance will be unable to commit itself to good governance in the future.
While people usually think of governments as a single entity, it is more analytically useful to consider governments as a complex network between various special interest groups (including government bureaucracies). These special interest groups support the existing government in return for rents, for example, monopoly privileges, handouts, and the restriction of competition. Improving governance by removing these rents will cause the interest group which receives them to fight back. For example, Mexican teachers went on strike when a law was passed to ban them from selling their teaching positions.
Even if we assume no special interest groups want to block governance improvements, there remains the problem of the implementation. For example, if a new agency is created to fight corruption, who will police that agency? Los Zetas, a Mexican cartel, were originally commandos from the Mexican Army. In the 90’s Los Zetas decided the cartels paid better and defected, joining the Gulf Cartel.
Free cities can help overcome both of the above difficulties of improving governance. First, by building on unoccupied land, a free city can avoid battles with special interest groups. Second, by bringing in a group dedicated to economic growth, a free city can ensure fast and efficient implementation.
There are two types of free cities, charter cities and proprietary cities. Charter cities are the brainchild of Paul Romer and governed by developed countries. Proprietary cities is a term I use to describe private cities with institutional autonomy. The two ideas are similar in that both are a mechanism to introduce good governance to jump-start economic growth. The ideas differ in which body would be responsible for governance.
Developed countries have histories of good governance, making them a trustworthy steward of governance. Private companies, on the other hand, have less history, but if they own the land on which the city is built they will have an incentive to promote economic development to increase the value of the land. Private governance is also more widespread than commonly realized, dominated international trade dispute resolution for example.
One of the advantages of free cities over alternative effective altruism causes is that free cities, once started, can pay for themselves. Building a free city would not be charity, rather it would be investment. The taxes raised in a successful free city would easily cover the initial investment.
Further, free cities can improve governance in their host country outside the city itself. China, after seeing the success of Hong Kong, created four special economic zones. After those zones were successful, more were created. Hong Kong and then Shenzhen not only lifted their residents out of poverty. They showed the rest of China the impact governance improvement could have. Those governance improvements were then adapted by the rest of China, creating the Chinese economy we know today.
Lastly, free cities are viable today. The primary barrier is not technological, but rather political. Countries are unwilling to give up authority. Even though free cities would be built on uninhabited land. Choosing to live in a free city would be strictly voluntary.
Luckily, political barriers are slowly falling. Madagascar seriously considered building a charter city. However, Marc Ravalomanana, the president was ousted from power before Romer’s vision could be implemented. Honduras created the legal framework to create free cities, calling them ZEDEs (zonas de empleo y desarollo economico). However, implementation of the legal framework has stalled.
Private capital also exists for the construction of a free city. King Abdullah Economic City, Saudi Arabia, the first city to be listed on a stock exchange is estimated to cost $86 billion. Songdo International Business District, Korea, cost an estimated $40 billion.
Of course it is possible free cities will fail, any number of things could go wrong. On the other hand, sustained economic growth is the best anti-poverty tool humans have ever known. Free cities could lift tens of millions, if not hundreds of millions, of people out of poverty in the next century by improving governance. For this reason, I believe effective altruists should add free cities to their pantheon of causes.
Two recent articles illustrate the Silicon Valley approach to building cities. The first is by Eliezer Yudkowsky, which he wrote as a response to Y Combinator’s city project. The second is by Yury Lifshits.
The most interesting part of both articles is the discussion of creating a special economic zone for new cities. Yudkowsky writes, “whole city that negotiated a special economic zone with some state or country.” Though he is suspects the host country will violate any agreement to create a special economic zone if the new city is successful. Lifshits writes,
Neocities are more likely to be developed as privately managed, special economic zones as opposed to traditional cities led by democratically elected officials. A neocity model should include a high-level plan for governance structure and key policy innovations.
He takes it as a given that new cities should have a degree of legal autonomy from the host country.
One thing missing from both discussions is what might be called base law. Will the zone use common law or civil law? Will private arbitration agreements be respected by the courts? Given both articles imply the cities will be built in developed countries, there is an implicit assumption of functioning courts. Nevertheless, there should be discussion around how to improve commercial law.
The second missing aspect is technology regulations. I am a bit surprised there is no discussion in either piece. A new city could allow experimentation with drone technology, for example, which is currently prohibited under US law. Given the difficulty of attracting people to a new city in urbanized countries, creating a regulatory environment to attract tech innovation seems an obvious step.
A third thing to take note of is city location. The vast majority of urbanization is occurring in the developing world. It would be a shame if Silicon Valley only focuses on developed countries.
A private company is proposing to build eight new cities in Australia. The cities would be linked by high speed rail with Sydney, Melbourne, and Canberra. The estimated investment is $150 billion U.S. It will be recouped by the increase in land prices one the necessary infrastructure is in place. Part of the article is excerpted below.
The most novel aspect of the plan, however, is its financing. CLARA proposes to build the line entirely with private funds. A News.com.au report puts the total price tag at A$200 billion ($150.6 billion U.S.); the company proposes to pay this sum back to its lenders by capturing some of the increase in the value of the land surrounding the rail stations, with revenue from increases in land values above that point returning a profit to the investors, who are currently located in Australia and the United States. CLARA chairman Nick Cleary told News.com.au that land the company purchased at $A1000 ($750 U.S.) per lot could be sold for as much as A$150,000 ($113,000 U.S.) once the necessary infrastructure is in place.
I have a humble suggestion, ask the Australian government to create a special economic zone for the new cities. I am not terribly familiar with Australian governance, but I am sure there are improvements to be made. Lower taxes, simplified business registration, and expedited construction permits could all cut the cost of building the new city. Perhaps there are regulations on technology which are slowing innovation in Australia, the new cities could ask to opt out of those regulations. The scale of the project makes it reasonable to ask for a certain amount of autonomy. In fact, it would be foolish not to.