Why Silicon Valley and not Silicon Kansas? The soil in Kansas is richer. The land is cheaper by an order of magnitude. The people work just as hard. Yet a one-bedroom apartment in San Jose costs more than a farmhouse on 40 acres outside Wichita, and the economic output per square mile in Santa Clara County exceeds that of entire midwestern states. Something about economic geography - the study of where economic activity happens and why it clusters there - explains this puzzle. And once you see the pattern, you start noticing it everywhere: why London dominates European finance, why Shenzhen builds your electronics, why a fishing village in the Persian Gulf became Dubai, and why some resource-rich nations stay desperately poor while resource-barren ones get rich.
Geography doesn't just provide the stage for economic activity. It writes half the script. The distance between a mine and a port, the depth of a harbor, the proximity of a university to a startup incubator, the accident of sitting on top of an oil field or along a river navigable by barges - these physical and spatial facts shape trillions of dollars in economic outcomes every year. Ignore geography and economics becomes a fantasy played out on a featureless plane. Factor it back in, and suddenly the wealth gaps between nations, regions, and even neighborhoods start making a brutal kind of sense.
Agglomeration: Why Winners Keep Winning
The single most powerful concept in economic geography has an unglamorous name: agglomeration. It means clustering. Firms in the same industry tend to bunch together, and once they start bunching, the bunch grows. Hollywood didn't become the film capital because California has better light than Georgia (it barely does). It became Hollywood because once a few studios set up there in the 1910s, actors moved nearby, prop shops opened, talent agencies followed, film schools launched, and within two decades an entire ecosystem existed that no other city could replicate without starting from scratch.
Alfred Marshall identified this dynamic back in 1890, pointing to three forces that pull firms together. A concentrated labor pool means companies can hire specialized workers without relocating them. Shared suppliers reduce costs because a circuit board manufacturer serving 50 tech firms can achieve economies of scale impossible if those firms were scattered across 50 different cities. And knowledge spillovers - the hardest force to measure but arguably the most potent - mean ideas leak between firms through hallway conversations, job-hopping employees, and after-work meetups at the same three bars.
Economists at Harvard studied patent citations and found that inventors are dramatically more likely to cite other inventors located within 25 miles. Not because they read each other's papers - because they talk at conferences, run into each other at cafes, and hire each other's former students. Physical proximity accelerates innovation in ways that Zoom calls still cannot fully replicate. This is why tech companies spent billions on gleaming campus headquarters even after the pandemic proved remote work was technically feasible.
Silicon Valley is the textbook agglomeration story. William Shockley brought the transistor to Mountain View in 1956 because he wanted to live near his mother in Palo Alto. That personal choice snowballed. Eight of his employees left to found Fairchild Semiconductor. People from Fairchild spun off Intel, AMD, and dozens of others. Stanford University fed the cluster with talent and research. Venture capital firms set up on Sand Hill Road to be close to the deal flow. By the 1990s, the Valley's gravitational pull was so strong that startups moved there not because the geography made sense, but because being anywhere else meant missing out on the network.
The same logic explains Wall Street in Manhattan, the pharmaceutical corridor along the New Jersey Turnpike, Bollywood in Mumbai, watchmaking in the Swiss Jura, and the ceramic tile industry concentrated in Sassuolo, Italy - a town of 40,000 that produces more tiles than most countries. Once a cluster hits critical mass, it becomes self-reinforcing. That is both the promise and the problem of agglomeration: it creates extraordinary productivity, but it also concentrates wealth geographically, hollowing out places that lose the clustering lottery.
Location Theory and the Gravity of Markets
Long before anyone talked about tech hubs, economists tried to crack a simpler question: where should a factory go? Johann Heinrich von Thunen modeled it in 1826 with concentric rings around a city - perishable goods produced close, durable goods farther away, forestry in between. His model was crude (it assumed flat, featureless terrain and a single market), but the core insight endures: transportation costs shape what gets produced where.
Alfred Weber sharpened this in 1909 by asking whether a factory should locate near its raw materials or near its customers. The answer depends on what economists call the material index - the ratio of raw material weight to finished product weight. Steel mills locate near iron ore and coal because smelting sheds enormous weight; you don't want to haul tons of slag. Bakeries locate near customers because flour gains weight when you add water and bake it. Soft drink bottling plants sit near markets because shipping water across a continent is absurd when every city already has water.
When Toyota chose Georgetown, Kentucky for a major assembly plant in 1986, the decision wasn't random. Georgetown sits within a day's truck drive of roughly two-thirds of the US population. Land was cheap. Labor was available. Kentucky offered $147 million in incentives. But the clincher was geographic centrality - minimizing the average shipping distance to dealers from coast to coast. That single factory now produces 550,000 vehicles per year, and over 100 Japanese-affiliated suppliers subsequently opened within a 200-mile radius. Weber's location triangle, playing out in real time with real billions.
Walter Christaller's central place theory from 1933 tackled why some cities become regional hubs while others stay small. His logic was geometric: basic services (grocery stores, gas stations) need only a small customer base, so they appear in every town. Specialized services (heart surgeons, opera houses, international airports) need huge customer bases, so they concentrate in larger cities spaced farther apart. The result is a hierarchy - villages serve hamlets, towns serve villages, cities serve towns, metropolises serve cities. It's an idealized pattern, but if you look at a map of settlements in the American Midwest or the German plains, the spacing is eerily close to what Christaller predicted.
Modern location decisions fold in variables Thunen and Weber never imagined: broadband speeds, time zone alignment with trading partners, climate appeal for recruiting talent, regulatory friendliness, and proximity to airports with direct flights to key markets. But the underlying question hasn't changed. Where you are still determines what you can do profitably, and how far you are from customers, suppliers, and talent still shapes the math of every business plan.
Special Economic Zones: Geography Hacked by Policy
If location determines economic destiny, what happens when a government tries to rewrite that destiny by fiat? The answer is the special economic zone (SEZ) - a geographically bounded area where the normal rules don't apply. Lower taxes, relaxed regulations, streamlined customs, sometimes even different labor laws. The idea is to create an artificial agglomeration advantage where geography alone didn't provide one.
5,400+ — Special economic zones operating worldwide across 147 countries as of 2024
China's Shenzhen is the most spectacular success story in the history of SEZs. In 1979, it was a fishing village of 30,000 people across the border from Hong Kong. Deng Xiaoping designated it one of four special economic zones, offering tax breaks and regulatory freedom to foreign investors. By 2024, Shenzhen's GDP exceeded $500 billion - larger than Sweden's - and its population had swelled past 17 million. The city now houses Huawei, Tencent, BYD, and DJI. It manufactures roughly 90% of the world's consumer electronics. A fishing village became a global technology capital in 45 years because policy rewrote the economic geography.
But for every Shenzhen, there are dozens of ghost zones. India created over 400 SEZs after 2005, yet many sit half-empty, plagued by land acquisition disputes and infrastructure gaps. Honduras tried "model cities" with near-sovereign autonomy - the experiment collapsed amid constitutional challenges and accusations of neo-colonialism. Russia's SEZs have produced mixed results at best. The lesson from economic geography is uncomfortable: you can bend location advantages with policy, but you can't ignore the underlying spatial logic. An SEZ with no port access, no highway, and no nearby labor force will fail no matter how generous the tax breaks.
The Resource Curse: When Geographic Wealth Breeds Economic Poverty
Here's a paradox that should bother you. The Democratic Republic of Congo sits on an estimated $24 trillion in mineral wealth - cobalt, coltan, diamonds, gold, copper, tin. It is one of the poorest countries on Earth. Nigeria is Africa's largest oil producer. Over half its population lives on less than $2.15 a day. Venezuela has the planet's largest proven oil reserves. Its economy contracted by 75% between 2013 and 2021. Meanwhile, resource-poor Japan, Singapore, and South Korea built some of the richest societies in human history.
Economists call this the resource curse (or the "paradox of plenty"), and it's one of the most well-documented phenomena in economic geography. The mechanism works through several reinforcing channels. Oil and mineral revenues flood the government with money that doesn't come from taxing citizens, which weakens the social contract - why build accountable institutions when the treasury fills itself from the ground? The export revenue inflates the currency, making non-resource industries uncompetitive on global markets (a dynamic called Dutch disease, named after the Netherlands' experience when North Sea gas revenues crippled Dutch manufacturing in the 1960s). Political elites fight over resource rents instead of investing in education and infrastructure. And the volatility of commodity prices whipsaws the entire economy - when oil goes from $100 to $30, a petrostate's budget doesn't just shrink, it collapses.
Nigeria: World's 6th-largest oil producer; 63% poverty rate; oil represents 90% of exports but only 10% of GDP due to enclave economics
DRC: $24T in mineral reserves; GDP per capita of $580; decades of conflict over resource control
Venezuela: 303 billion barrels of proven reserves; hyperinflation hit 130,000% in 2018; 7.7 million people fled the country
Norway: Major oil producer; sovereign wealth fund worth $1.6 trillion; invests revenue abroad to avoid Dutch disease
Botswana: Diamond wealth; invested in education and governance; grew from one of the world's poorest nations to upper-middle income in 50 years
Chile: World's largest copper exporter; stabilization fund smooths commodity price swings; diversified economy
Norway's escape from the curse is instructive. When North Sea oil started flowing in the 1970s, Norway established the Government Pension Fund Global - now worth over $1.6 trillion - which invests oil revenues abroad rather than spending them domestically. This prevents currency appreciation, forces the government to budget from tax revenue like any normal economy, and saves wealth for future generations. It sounds simple. Almost no other petrostate has managed it, because the political incentives to spend windfall revenue right now are overwhelming.
The geographic dimension matters profoundly here. Resources are where they are. You can't move an oil field to a country with better institutions. So the question becomes whether institutions can be built strong enough to manage the geographic lottery of sitting on valuable commodities. Botswana did it with diamonds. Chile has largely done it with copper. But the list of failures is much longer, and the pattern is clear enough that the International Monetary Fund now specifically advises resource-rich developing nations on institutional design to avoid the curse.
Trade Corridors: The Arteries of Global Commerce
Every product you touch today traveled a geographic path to reach you. Your phone's cobalt came from the DRC, was refined in China, assembled in a Foxconn factory, shipped through the Strait of Malacca, crossed the Pacific, cleared customs in Long Beach, and rode a truck to your local store. That chain of movement follows trade corridors - geographic routes where infrastructure, policy, and physical geography converge to create channels of commerce.
Some corridors are dictated by pure geography. The Strait of Malacca, a narrow waterway between Malaysia and Indonesia, handles roughly 25% of all global maritime trade because it's the shortest route between the Indian Ocean and the Pacific. The Suez Canal slices through Egypt to save ships from circumnavigating Africa - when the container ship Ever Given blocked it for six days in March 2021, it held up an estimated $9.6 billion in trade per day. The Panama Canal does the same for the Americas. These chokepoints are where geography grabs global commerce by the throat.
China's Belt and Road Initiative (BRI), launched in 2013, is the most ambitious attempt in modern history to engineer new trade corridors. With over $1 trillion in infrastructure investments across 140+ countries, the BRI builds ports, railways, highways, and pipelines connecting China to markets in Southeast Asia, Central Asia, Africa, and Europe. The China-Pakistan Economic Corridor alone represents $62 billion in investment, including a deep-water port at Gwadar that gives China access to the Arabian Sea without passing through the Strait of Malacca. It's economic geography as grand strategy - rewriting the map of global trade routes through sheer infrastructure spending.
The old Silk Road worked the same way. For over a millennium, goods moved between China and the Mediterranean along a corridor shaped by mountain passes, oasis towns, and the gaps between deserts. Samarkand, Bukhara, and Kashgar became wealthy not because of what they produced, but because of where they sat. Position along a corridor can be worth more than any natural resource - a lesson Singapore understands better than almost any nation on Earth. Sitting at the crossroads of Asian maritime trade, Singapore has no natural resources, no agricultural land, and no domestic market to speak of. Its GDP per capita exceeds $80,000.
Regional Inequality: The Geography of Haves and Have-Nots
Zoom out from individual clusters and corridors, and a starker picture emerges: economic activity is not distributed evenly across any country, and the gaps are widening in most of them. This is regional inequality, and it's one of the defining political and economic challenges of the 21st century.
The numbers are staggering. Half of South Korea's economic output comes from the Seoul metropolitan area. London and its immediate surrounds generate 40% of the United Kingdom's GDP. In many developing countries, the ratio is even more extreme - Lima produces nearly half of Peru's GDP, Bangkok about 44% of Thailand's. The pattern holds almost universally: a small number of cities and regions generate a disproportionate share of national wealth, while vast areas are left behind.
This isn't just an economic abstraction. It reshapes lives. A young person born in the north of England earns, on average, 15-20% less than someone born in London over their lifetime - not because of talent differences, but because of where the jobs, networks, and investment concentrate. In the United States, the divergence between "superstar cities" (New York, San Francisco, Seattle, Boston) and the rest of the country has accelerated since the 1980s. Economist Raj Chetty's research shows that the zip code a child is born into predicts their adult income better than almost any other single variable.
Regional inequality is not just an economics problem - it's a destabilizing political force. Brexit was driven disproportionately by voters in deindustrialized regions of England that felt abandoned by London-centric prosperity. The rise of populist movements in the American Rust Belt, France's gilets jaunes protests, and Italy's persistent North-South divide all trace back to geographic concentrations of economic opportunity. When a nation's wealth clusters in a handful of cities while other regions stagnate, the political consequences can be seismic.
Why does this happen? Agglomeration, again. The same forces that pull firms and workers into clusters also drain them from everywhere else. When a talented graduate in Appalachia moves to New York for a better job, Appalachia loses a taxpayer, a potential entrepreneur, and a community member. New York gains all three. Economists call this cumulative causation - success breeds the conditions for more success, while decline breeds the conditions for more decline. Gunnar Myrdal described these as "backwash effects" and "spread effects." In theory, prosperity should eventually spread outward from core regions. In practice, the backwash often dominates, and the spread never arrives.
The New Economic Geography: Krugman's Revolution
For most of the 20th century, mainstream economics largely ignored space. Models assumed goods moved freely, workers relocated instantly, and geography was irrelevant. Then Paul Krugman published a paper in 1991 called "Increasing Returns and Economic Geography" that cracked the field wide open. He won the Nobel Prize for it in 2008.
Krugman's insight was deceptively simple: combine increasing returns to scale, transportation costs, and the ability of workers to move, and you get a model where economic activity naturally concentrates in a few locations even when the underlying geography is identical everywhere. The math showed that beyond a certain threshold of agglomeration, the process becomes self-sustaining. Workers move to where the jobs are, firms move to where the workers are, and the loop accelerates until you get a core-periphery pattern - dense, productive cores surrounded by thinner, dependent peripheries.
The takeaway: Geography doesn't just happen to economies - it shapes them through self-reinforcing feedback loops. Once a region gains an edge in skilled labor, suppliers, and knowledge networks, that advantage compounds. Krugman's models showed this mathematically, but any resident of a dying factory town or a booming tech hub already knew it intuitively.
The New Economic Geography (NEG) framework explains why globalization hasn't flattened the world the way some predicted. Thomas Friedman's 2005 claim that "the world is flat" - that technology and trade had eliminated the advantages of location - turned out to be spectacularly wrong. If anything, globalization has made geography more important, not less. It increased the rewards for being in the right place (a globally connected city) and the penalties for being in the wrong one (an isolated rural area with no broadband and no airport). The internet lets a programmer in rural Montana theoretically compete with one in San Francisco. In practice, the San Francisco programmer has access to venture capital meetings, industry conferences, and a peer network that the Montana programmer simply does not.
Global Value Chains: Where Geography Meets Production
Modern production doesn't happen in one place. An iPhone contains components from over 40 countries. A Boeing 787 Dreamliner's parts are manufactured on four continents before final assembly in Everett, Washington. This fragmentation of production across space is what economists call global value chains (GVCs), and they represent one of the most important developments in economic geography over the past 30 years.
The logic is straightforward. Different stages of production have different geographic requirements. Research and design cluster in places with highly educated workforces and strong intellectual property protections - California, Bavaria, South Korea. Component manufacturing goes where precision engineering is available at competitive costs - Taiwan for semiconductors, Japan for specialty materials, Germany for machine tools. Assembly gravitates toward large, disciplined labor forces with low wages and good logistics - Vietnam, Bangladesh, Mexico. Marketing and management stay near major consumer markets. The result is a geographic chain where each link occupies the location best suited to its particular task.
(high-income hubs)
(specialized industrial regions)
(low-cost labor markets)
(near consumer markets)
The geographic implications are enormous. Countries don't need to build entire industries from scratch anymore - they can plug into existing value chains at the stage matching their comparative advantage. Vietnam went from virtually no electronics industry in 2000 to exporting $114 billion in electronics by 2023, largely by becoming Samsung's primary assembly location. Bangladesh did something similar with garments. Ethiopia was trying to do it with shoe and clothing manufacturing before political instability disrupted the trajectory.
But value chains also create geographic vulnerabilities. When Thailand flooded in 2011, hard drive production worldwide dropped 40% because so much of the global supply was concentrated in flooded Thai factories. The COVID-19 pandemic exposed similar fragilities - semiconductor shortages cascaded through automobile, electronics, and appliance production because a handful of fabrication plants in Taiwan and South Korea represented irreplaceable bottlenecks. The geographic concentration that makes value chains efficient also makes them brittle. This tension between efficiency and resilience is now driving a partial reorganization of global production patterns, with concepts like "friendshoring" and "nearshoring" reflecting a new awareness that trade geography carries strategic risk.
Development Geography: Why Poor Countries Stay Poor
No question in economic geography carries more moral weight than this one: why are some countries rich and others poor? The map of global wealth is not random. It clusters. Sub-Saharan Africa, South Asia, and parts of Central America are persistently poor. North America, Western Europe, East Asia, and Oceania are persistently rich. That pattern has been remarkably stable for decades despite trillions of dollars in development aid, and geography is a central part of the explanation.
Jeffrey Sachs has argued that tropical geography itself imposes economic penalties. Tropical diseases (malaria alone costs sub-Saharan Africa an estimated $12 billion per year in lost productivity and treatment), poor soil quality in tropical regions, vulnerability to droughts and floods, and distance from major world markets all create structural disadvantages that no amount of good governance can fully overcome without targeted investment. A landlocked tropical country faces compounding geographic penalties - no port access means higher trade costs, which means less foreign investment, which means slower growth, which means less money to build the infrastructure that might offset the geographic disadvantage.
Daron Acemoglu and James Robinson pushed back with an institutional argument. In their influential work, they contend that geography matters, but primarily through historical channels. Colonial powers set up extractive institutions in places with tropical diseases (where European settlers died quickly) and inclusive institutions in temperate climates (where settlers planned to stay). Those institutional legacies persisted long after colonialism ended. Botswana and the DRC have similar latitudes but radically different institutions - and radically different economic outcomes.
The truth almost certainly involves both. Geography sets the initial conditions. Institutions determine how effectively a society responds to those conditions. A landlocked country with strong institutions (like Switzerland) can overcome geographic disadvantages. A coastal country with weak institutions (like Somalia, which has one of Africa's longest coastlines) can squander geographic advantages. But pretending geography doesn't matter is as naive as pretending institutions don't matter. The interplay between physical space and human organization is where economic development actually happens, and untangling the two is one of the hardest problems in development economics.
Energy Geography and the Spatial Logic of Power
The map of energy resources has shaped geopolitics for over a century, but it's also profoundly shaped economic geography. Where energy is produced, how it moves, and who controls the chokepoints - these factors determine which regions industrialize, which cities grow, and which nations wield outsized global influence.
The fossil fuel era concentrated economic power in a few geographic zones. The Persian Gulf holds roughly 48% of proven global oil reserves. Russia and the Caspian region add another 15%. This geographic concentration created petrostate economies, funded sovereign wealth funds worth trillions, and drew military alliances and conflicts that reshaped maps. The Strait of Hormuz, through which 21% of global oil passes daily, is so strategically vital that the US Fifth Fleet is permanently stationed in Bahrain specifically to keep it open.
The energy transition is beginning to redraw this map. Solar energy favors the Sahara, the Arabian Peninsula, the American Southwest, and Australia - regions with high solar irradiance. Wind energy concentrates along coastlines, mountain ridges, and the Great Plains. But the geography of critical minerals needed for batteries, solar panels, and electric motors creates new concentrations. The DRC produces 70% of the world's cobalt. China refines 60% of global lithium and manufactures 80% of solar panels. Chile, Argentina, and Bolivia hold most of the world's lithium reserves in what's been called the "lithium triangle." The geographic curse of fossil fuels may simply be replaced by the geographic curse of transition minerals.
Urban Economic Geography: Why Cities Eat the World
Cities occupy about 3% of the Earth's land surface. They generate over 80% of global GDP. That ratio - 3% of space producing 80% of value - captures the central fact of urban economic geography: cities are the most productive spatial arrangement humans have ever invented, and they keep getting more dominant.
80% — of global GDP produced on just 3% of Earth's land surface - in cities
Edward Glaeser's research has documented the productivity premium of urban density. Workers in cities earn more, not just because city jobs pay better, but because cities make workers more productive. The mechanism is the same agglomeration logic scaled to the urban level: denser networks of contacts, faster knowledge transfer, more specialized labor markets, and thicker supplier ecosystems. A lawyer in New York can specialize in Uzbek trade law because there are enough clients in New York to sustain that niche. A lawyer in a town of 10,000 handles everything from wills to traffic tickets because specialization isn't viable.
This productivity advantage explains why urbanization has been the dominant migration trend for two centuries and shows no sign of reversing. Over 56% of the world's population now lives in urban areas, projected to reach 68% by 2050. In high-income countries, urbanization already exceeds 80%. The economic logic is simple and relentless: cities offer better jobs, higher wages, more services, and more opportunities for advancement. The costs - housing prices, congestion, pollution, social isolation - are real but evidently not enough to reverse the flow.
The geography within cities matters enormously too. Urban economists have shown that proximity to a city center typically correlates with higher land values, wages, and productivity, falling off with distance in a pattern called the urban wage gradient. Neighborhoods within the same city can have wildly different economic trajectories depending on transit access, historical investment, zoning laws, and the self-reinforcing dynamics of reputation. The difference in median household income between the richest and poorest neighborhoods in New York City exceeds a factor of 10. Geography operates at every scale, from the global to the block level.
Agriculture and the Geography of Food Production
Before there were tech hubs and financial centers, economic geography was fundamentally about where food could grow. It still is, more than most people realize. Agricultural output depends on climate zones, soil quality, water availability, and terrain in ways that no amount of technology has fully overcome. The American Midwest produces a third of the world's corn and soybeans because of a unique combination of deep prairie topsoil, adequate rainfall, flat terrain for mechanized farming, and a river system (the Mississippi) that moves grain to export terminals at minimal cost.
The Green Revolution of the 1960s and 1970s boosted yields dramatically through high-yield crop varieties, synthetic fertilizers, and irrigation, but it did so unevenly across geographies. Regions with existing irrigation infrastructure and relatively flat terrain (the Punjab in India, the Central Valley in California, the Nile Delta) benefited enormously. Hilly, rain-dependent regions in sub-Saharan Africa benefited far less. The result was a geographic divergence in agricultural productivity that persists today. Corn yields in Iowa average over 11,000 kilograms per hectare. In Malawi, they average around 2,000. Same crop, different geography, fivefold difference.
Climate change is now reshuffling this geographic deck. Traditional breadbaskets face rising temperatures, shifting rainfall patterns, and more frequent extreme weather events. Meanwhile, previously marginal lands in northern Canada, Scandinavia, and Siberia are becoming viable for agriculture as frost-free seasons lengthen. The economic geography of food production in 2050 will look meaningfully different from today's map, with enormous implications for land values, migration patterns, trade flows, and geopolitical power.
Digital Economy: Does Geography Still Matter?
The rise of the internet prompted widespread predictions that geography would become irrelevant. If you can transmit information at the speed of light, why does it matter where a company sits? Thirty years into the commercial internet, the answer is clear: geography matters more than ever in the digital economy, just in different ways.
Data centers - the physical infrastructure of the cloud - cluster in specific geographic locations dictated by power costs, cooling requirements, network connectivity, and regulatory environments. Northern Virginia hosts the largest concentration of data centers on Earth, partly because of cheap electricity and proximity to major internet exchange points. Iceland and the Nordic countries attract data centers with cheap geothermal and hydroelectric power plus natural cooling from cold climates. Even the supposedly "placeless" digital economy has a very specific geography.
High-frequency trading firms in financial markets pay millions of dollars for offices physically closer to stock exchange servers, because the speed of light creates measurable latency differences over distance. A firm trading from a server farm in New Jersey - mere miles from the NYSE data center - has a 13-millisecond advantage over one trading from Chicago. In markets where billions change hands in microseconds, 13 milliseconds is a fortune. The speed of light is a geographic constraint, and geography is profit.
The tech industry itself defied the "death of distance" prophecy. Rather than dispersing, it concentrated more intensely than almost any industry in history. Five metropolitan areas (San Francisco, Seattle, New York, Boston, and Austin) captured 90% of US tech job growth between 2005 and 2017. Remote work expanded during the pandemic, but the highest-paying tech jobs and venture capital investment remain overwhelmingly concentrated in a handful of cities. The network effects that drive agglomeration in physical industries turn out to be even stronger in knowledge-intensive digital industries, where face-to-face interaction accelerates the kind of tacit knowledge transfer that drives innovation.
Even for digital workers who can theoretically work from anywhere, geography shapes their economic outcomes. Remote workers in low-cost cities earn less than their counterparts in high-cost cities doing identical work, because companies increasingly adjust compensation to local cost of living. The geographic wage premium has proven remarkably persistent even when the technical requirement for geographic proximity has evaporated. Place still pays.
Putting the Map Together
Economic geography isn't one story. It's a layered accumulation of forces - physical geography setting the initial conditions, historical trade routes and colonial legacies shaping institutions, agglomeration pulling economic activity into clusters, policy interventions like SEZs trying to redirect the flow, global value chains fragmenting production across space, and technology simultaneously reinforcing and occasionally disrupting established patterns.
The field's central insight is that economic outcomes are not geographically random, and they're not geographically fixed. They're the product of feedback loops between place and human activity that compound over time. Detroit didn't decline because Michigan moved. It declined because the automotive industry restructured, skilled workers left, the tax base eroded, infrastructure deteriorated, and each wave of departure made the next one more likely. Shenzhen didn't rise because Guangdong Province suddenly acquired new natural resources. It rose because policy, investment, and migration created an agglomeration effect that fed on itself for four decades.
Understanding these dynamics matters for anyone making decisions about where to live, where to start a business, where to invest, or how to vote on policies affecting regional development. The zip code question - why does where you are shape what you can become? - is not a thought experiment. It's the lived reality of billions of people whose economic opportunities are expanded or constrained by the geography they inhabit. And as climate change reshuffles physical geography, as the energy transition rewires the resource map, and as digital technology continues to both concentrate and occasionally disperse economic activity, the spatial logic of the economy will keep evolving. Economic geography isn't a settled map. It's a map that keeps redrawing itself, with real consequences for every point on it.