EOSE09 Literature Notes: Models of Market Integration & Regional Growth - PDF

EOSE09 - literature notes Lecture 2 - Models of market integration and regional growth Robert J. Barro & Xavier Sala-i-Martin (1991). Convergence across States and Regions Links to an external site., Brookings Papers on Economic Activity, Economic Studies Program, The Brookings Institution, vol. 2...

EOSE09 - literature notes Lecture 2 - Models of market integration and regional growth Robert J. Barro & Xavier Sala-i-Martin (1991). Convergence across States and Regions Links to an external site., Brookings Papers on Economic Activity, Economic Studies Program, The Brookings Institution, vol. 22(1), pages 107-182. Williamson, J.G. (1965) Regional inequality and the process of national development: Links to an external site. A description of the patterns.Links to an external site. Economic Development and Cultural Change 13(4): 3-45. ​ What is the big question? Barro & Sala-i-Martin: Do economies or regions with lower initial incomes grow faster than those with higher initial incomes, leading to income convergence? Williamson: How does regional inequality evolve over the course of a country’s national development, and what patterns can be observed in this process? ​ How are the authors trying to answer that question (methods) Barro & Sala-i-Martin: The authors employ empirical analysis, focusing on the concept of "β-convergence" (growth rates inversely related to initial income levels) and "σ-convergence" (a reduction in income dispersion over time). They analyze cross-sectional and panel data from U.S. states, European regions, and other global datasets. They use regression models to test the speed and existence of income convergence. Williamson: Williamson uses descriptive and comparative analysis, examining historical and cross-country data to observe patterns in regional income inequality during different stages of development. He identifies key phases in national development and correlates them with inequality trends. ​ What are the results of their analysis Barro & Sala-i-Martin: The analysis supports the existence of β-convergence across regions and states, indicating that poorer areas grow faster than richer ones, controlling for other factors such as education and savings. However, the rate of convergence is slow, and absolute convergence is not always observed without conditioning on structural factors. Williamson: Williamson finds that regional inequality initially increases during the early stages of industrialization as development is concentrated in certain regions. Over time, inequality tends to decline as development spreads to lagging regions. This is depicted as an inverted-U curve of regional inequality over time. ​ Which factor does Barro / Sala-i-Martin list as most important for understanding growth. How does that contrast with Williamson? Barro & Sala-i-Martin: investment in human capital (education and skills development) as a critical driver of growth and convergence. Williamson: Williamson highlights industrialization and structural transformation as the primary drivers of growth and regional inequality dynamics. Lecture 3 - Historical evidence, convergence and the creation of factor markets Compare Sweden and Italy based on these factors. -​ The pattern of long-term regional inequality, how do these countries differ? Sweden: Experienced a steady decline in regional inequality from 1860 to 1980, without following the typical inverted U-shaped pattern seen in other countries. However, since 1980, regional disparities have slightly re-emerged due to the concentration of knowledge-based industries in metropolitan areas. Italy: Followed a classic inverted U-shaped pattern, with regional inequality rising until the mid-20th century and then partially declining. However, convergence stalled after the 1970s, leaving persistent disparities between the industrialized North and the underdeveloped South (Mezzogiorno). -​ List all explanatory factors for these patterns that you find in the texts.  -​ Specifically, pay attention to the role of state policies. How have they been different? Sweden: ​ From 1940 to 1980, state policies played a major role in further reducing regional inequalities. ​ Government intervention promoted labor reallocation from declining to growing sectors, ensuring a smoother economic transition. ​ Strong institutional support and education policies helped equalize opportunities across regions. Italy: ​ State intervention was significant but largely unsuccessful in the long run.  ​ From the 1950s to the 1980s, major regional policies attempted to industrialize the South, leading to temporary convergence. ​ However, these policies failed to create self-sustaining growth, and the economic gap persisted beyond the 1970s. ​ Weak governance, inefficient policy implementation, and low social capital in the South contributed to policy failure. Enflo, Kerstin & Joan Ramón Rosés (2015) Coping with regional inequality in Sweden: structural change, migrations, and policy, 1860–2000,Links to an external site.Economic History Review 68 (1) pp. 191-217 Short summary: -​ In most countries, regional income inequality follows an inverted U-shape, rising with industrialization and later declining. -​ Sweden, however, experienced a steady decline in inequality from 1860 to 1980, making it lower today than in other European countries. -​ The main driver of this long-term convergence was structural change, while neoclassical and technological factors were less influential. -​ The process occurred in three phases: 1. 1860–1940: Market expansion and high migration reduced regional income gaps. 2. 1940–1980: Government policies and institutional arrangements further equalized incomes. 3. 1980–2000: Regional divergence reappeared as migration slowed and knowledge-based industries concentrated in major cities. The research question: The paper investigates why Sweden's regional income inequality declined steadily from 1860 to 2000, contrasting with the inverted U-shaped pattern seen in other countries. The authors focus on regions rather than entire countries to analyze localized economic dynamics, structural change, and migration patterns that influenced regional convergence. Specific hypothesis: The authors examine: ​ Whether Sweden’s regional income inequality followed the standard inverted U-shape. ​ The role of structural change, migration, and government policies in reducing regional disparities. ​ The impact of market forces vs. institutional interventions on regional convergence. They also investigate if there was a convergence across Swedish countries in the long run by using 2 measures of regional income convergence: -​ The decline of income dispersion -​ Unconditional convergence, whether poor regions grew faster than richer regions Methods used in the article: The study uses historical data analysis, combining wage, GDP, and employment statistics across Swedish regions from 1860 to 2000. The authors apply quantitative methods to measure income disparities over time and assess the influence of migration, structural shifts, and policy changes on regional convergence. Results from the study: Sweden experienced continuous regional convergence from 1860 to 1980, followed by regional divergence after 1980. Conclusion: Sweden’s regional income inequality declined due to market expansion (1860–1940), policy-driven redistribution (1940–1980), and structural changes. However, after 1980, regional divergence emerged as knowledge-based industries concentrated in metropolitan areas. Differences: Unlike many countries, Sweden did not follow an inverted U-shaped curve. Instead, convergence was steady and largely policy-driven. Felice, Emanuele (2011) Regional value added in Italy, 1891–2001, and the foundation of a long‐term pictureLinks to an external site., Economic History Review 64 (3) pp. 929-950. Short summary: Felice (2011) reconstructs long-term regional GDP data for Italy from 1891 to 2001 to analyze the evolution of regional economic disparities. The study finds that Italy's regional inequality followed an inverted U-shaped pattern, with disparities rising until the mid-20th century and then partially declining. However, convergence slowed after the 1970s, leaving persistent gaps between northern and southern Italy. The research question: How have regional economic disparities in Italy evolved over the long term (1891–2001), and what factors have influenced their changes? Specific hypothesis: Italy’s regional inequality followed an inverted U-shape, increasing with industrialization and later declining. Structural changes, industrial policies, and institutional factors played key roles in shaping regional disparities. The Italian South (Mezzogiorno) faced persistent economic disadvantages due to weaker institutions, lower productivity, and slower industrialization. Methods used in the article: Reconstruction of regional GDP data over 110 years using historical national accounts. Analysis of regional value-added per capita in different sectors (agriculture, industry, services). Comparison of Italy’s regional disparities with other European countries. Results from the study: ​ 1891–1951: Regional disparities widened as the North industrialized faster than the South. ​ 1951–1971: Convergence occurred due to government intervention and industrial expansion in the South. ​ 1971–2001: Convergence slowed, and disparities persisted, with northern regions maintaining a strong economic lead. ​ Institutional and policy failures, particularly in the Mezzogiorno, contributed to long-term inequality. Felice concludes that while economic policies helped reduce disparities, they were insufficient to fully integrate the South into Italy’s industrial economy. The study highlights the need for institutional reforms and targeted policies to address long-term regional inequality. “Conditional convergence was at work; in other words, that a persistent negative conditioning variable (such as geographical position, culture, social capital, or even a mix of these factors) may have prevented the poorest regions from growing faster, but of course the specific variable remains unknown.” From the article: The north–south divide in Italy has been widely debated since the 19th century, with three main perspectives: (1) the traditional view that the North was naturally more suited for industrialization due to geography and human capital; (2) the dependency theory, arguing that the North exploited the South after Unification; and (3) the modern view, which sees regional differences as more complex, with internal variation within the South itself. Felice’s study presents new historical value-added estimates to clarify this debate. His findings indicate that the north–south divide emerged mainly during the interwar years (not immediately after Unification) and that southern Italy was already disadvantaged in literacy and life expectancy. He also examines the role of migration in economic convergence, highlighting that the South benefited less from early globalization. The study further explores the failure of state intervention in the South, which was temporarily successful in the 1950s–60s but later reversed. Lastly, it considers social capital as a key factor in regional disparities, aligning with recent research that links culture and economic performance. The article provides new long-term data on Italian regions, offering insights into industrialization, migration, policy, and economic inequality. Lecture 4 - The rise of cities Video of “Our Urban Age” Glaeser talks about how, in the 19th century, Americans moved into the vast continent, and spread out. In the 21st century we’re moving closer together and taking advantage of the benefits of being close to each other, and cities are growing tremendously. Think about the following questions: ​ What is it that makes cities productive? -​ Specialized in knowledge and creativity -​ Productive people want to go to cities “iron shapes iron” -​ Educated city has done better than non-educated cities -​ Skills make the individual better but also the people around the individual who take in each others skills -​ Long run success: talent and implementation of being an entrepreneur, easier when there are other entrepreneurs around who you can learn from -​ Big firms → more economic growth ​ Why is face-to-face interaction so crucial for growth? -​ Important for learning, seeing face to face if the information is taken in or not by the audience -​ Learn from smart people around you -​ Easier to focus when not online, and it could be harder to be entrepreneurial when only online ​ How have these advantages changed over time (especially lately)?  -​ Knowledge and skills becomes more and more important, cities also important as driver of growth -​ More expensive to live in the heart of a thriving city -​ Technological change → safer to share more stuff with each other (AirBnb, Uber for example) -​ Yelp can help know what to change and to measure development in areas in current time (Street view etc), gentrification Contrasting view: https://www.bloomberg.com/news/articles/2015-06-12/the-problem-of-urbanization-without-eco nomic-growth The problem of urbanization without economic growth -​ Today, rapid urbanization is not always accompanied by rising fortunes → particularly for the poorest people -​ They examine whether the urbanization of cities without an accompanying bump in living standards is linked to the growing megacities of today or if it is a part of a pattern across history -​ The authors study the connection between urbanization and economic development over the past five centuries, 1500 to 2010 -​ The study finds that urbanization without growth is not a new phenomenon. In fact, it recurs intermittently. -​ 1500: relatively weak relationship between urbanization and growth -​ 1950: relationship had changed considerably, much more strongly associated with each other. But the connection between urbanization and growth in this period was mainly a product of the most advanced nations. Developing and emerging nations remained rural and poor. -​ 2010: the relationship between urbanization and growth had changed again. Now poor countries, not the rich ones, are experiencing the most rapid urbanization. Overall, the researchers find that the relationship between urbanization and growth in this period is nearly identical to that of 1500: A tripling of GDP per capita is now associated with just a 13 percent increase in urbanization rates. By 1900 the mega-cities were concentrated in the most highly developed countries → by 1950 45% of the mega-cities were again in developing countries → by 2010 nearly 80% were in developing nations. Key takeaway: we should stop assuming that urbanization and development are inextricably linked. Rather it appears that urbanization is an innovation, not a mere consequence, of a more powerful economy. Glaeser, Edward L., and Matthew G. Resseger. "The complementarity between cities and skills. Links to an external site." Journal of Regional Science 50.1 (2010): 221-244. Short summary of the paper: - Urban economics suggests a strong link between city size and worker productivity, but this pattern only holds in highly educated cities in the U.S. - In the least educated metropolitan areas, city size has little impact on productivity or income, while in the most educated areas, population size explains 45% of productivity variation. - Key question: Why does city size boost productivity in skilled cities but not in unskilled ones? - Some of the productivity effect is due to skilled workers choosing bigger cities, but this only explains part of the relationship. Two main theories of agglomeration: 1. Non-knowledge-based (e.g., transport cost reduction, capital availability, geographic advantages). 2. Knowledge-based (e.g., learning from others, innovation, and knowledge spillovers). - The study explores whether urban density helps workers learn faster or accelerates innovation, affecting wage growth differently. - Findings suggest that workers in skilled cities learn faster, but wage growth patterns do not fully confirm whether learning or innovation is the main driver. Conclusion: If human capital keeps rising, cities will become even more important for economic growth, unless technology replaces the need for physical proximity. Conclusion from paper: This paper shows that agglomeration effects are much stronger in highly skilled cities, supporting knowledge-based theories over those focused on natural advantages or trade efficiency. However, direct evidence on knowledge-driven agglomeration remains limited. Replicating Glaeser and Mare (2001), we find that skill accumulation happens faster in metropolitan areas, especially in skilled ones. While this suggests a strong link between skills, city size, and learning, other tests find little conclusive evidence. A possible explanation is that skilled cities experience both faster learning and quicker technological change, creating mixed effects on wage growth. Further research is needed to clarify these dynamics. Jedwab, Remi and Dietrich Vollrath, 2015, Urbanization without growth in historical perspectiveLinks to an external site., Explorations in Economic History, Volume 58, Pages 1-21. Summary of Introduction Urbanization is often linked to economic development, but recent trends show that cities are growing even in poorer countries, a phenomenon called "urbanization without growth." This study examines the historical relationship between urbanization and income levels from 1500 to 2010. Before the mid-20th century, urbanization mainly occurred in wealthy, rapidly developing countries. Since then, urbanization has been concentrated in poorer nations, shifting the location of mega-cities from rich to poor countries. The paper explores the causes of this shift, distinguishing between positive urbanization (driven by improved infrastructure) and negative urbanization (driven by policy biases and rural poverty), with the latter playing a greater role in recent trends. Conclusion: Urbanization without growth, observed in the late 20th century, is part of a long-term trend dating back to 1500. Over the past five centuries, urbanization has increased at all income levels, rising by 25–30 percentage points. However, this process occurred in two phases: -​ before 1950, urbanization was concentrated in wealthy countries -​ while after 1950, it accelerated in poorer nations. → As a result, city size no longer reliably reflects living standards. This history shows that urbanization does not always mean industrialization or economic growth and can have both positive (e.g., lower urban mortality) and negative (e.g., rent-seeking) effects. Traditional theories linking urbanization to productivity shifts do not fully explain these patterns. Lecture 5 - Measuring growth, inequality and development The main text for this lecture is the introductory chapter (chapter 2) in the course book "Europe's regions". There are lots of maps, tables and graphs suplementing the text. Go through them and think about what they mean and whether these maps, tables and graphs support the conclusions drawn in the text. ​ Are they clear and informative? yes , most are but some are a bit hard to understand ​ What maps, tables and graphs support the arguments best, which are potentially redundant? redundant : figure 2.3 and table 2.5 (do not understand all the variables in the regression) ​ What are the "dangers of using modern definitions of regions and focusing on GDP" (Rosés and Wolf, Introduction). Rosés and Wolf warn that using modern regional definitions can misrepresent historical economic patterns, as borders and administrative units have changed over time. They also caution against focusing solely on GDP, which may overlook factors like living standards, inequality, and structural changes, leading to an incomplete understanding of regional development. ​ Think of alternative measures that are meaningful in explaining regional growth and development. -​ Productivity (output per worker) = reflects efficiency rather than just output -​ Human capital (education and skill levels) – captures long-term growth potential. To get a feeling for different for how long-run data can we used to inform about trend-breaks and potential causes also read the working paper by Miriam Fritzsche and Niko Wolf: ​ When do they observe an important trend break? They identify a significant trend break in the early 1960s. They explain that the shift from coal to oil during this period transformed regional coal abundance from an economic advantage into a disadvantage. This "oil invasion" led to a reversal of fortune for coal-dependent regions, as the new energy paradigm favored areas better adapted to oil-based industries.  ​ How do they explain it? They explain the trend break in the early 1960s as a result of the shift from coal to oil, which altered regional economic advantages. Previously, coal-rich regions had benefited from industrialization, but the widespread adoption of oil as the dominant energy source reduced coal’s economic importance. This transition disadvantaged coal-dependent regions, leading to economic decline, while regions better suited for oil-based industries gained a competitive edge. They argue that this shift was driven by technological changes, geopolitical factors, and market dynamics, which made oil a more efficient and flexible energy source. Literature: -​ Rosés, Joan and Nikolaus Wolf "Regional Economic Development in Europe, 1900-2010: A description of patterns", chapter 2 of The Economic Development of Europe's Regions: A Quantitative History Since 1900 -​ Fritzsche, Miriam, and Nikolaus Wolf. "Fickle Fossils. Economic Growth, Coal and the European Oil Invasion, 1900-2015.Links to an external site." (2023). CESifo Working paper 10805 Roses and Wolf: Chapter 2 of "The Economic Development of Europe's Regions: A Quantitative History since 1900," authored by Joan Ramón Rosés and Nikolaus Wolf, provides a comprehensive analysis of regional economic development patterns in Europe from 1900 to 2010. The chapter utilizes a newly constructed dataset encompassing regional employment structures and GDP per capita across European regions. The authors identify a U-shaped trajectory in regional income inequality: a decline from 1900 until around 1980, followed by an increase thereafter. They also observe a decrease in spatial coherence over the century, indicating that regional economic disparities became less geographically concentrated. The chapter emphasizes the significant role of industrial localization, particularly in Germany, eastern France, and northern Italy, and notes the historical dominance of capital regions in service sector concentration. The authors attribute these patterns to both short-term adjustments and long-term structural factors, including the impact of national institutional frameworks and sectoral employment shifts. Notes for the dugga Lecture 1 - why do we care about regions: -​ Andres Rodriguez-Pose (2018): → regions matter! argues that neglecting economically lagging regions has fueled political discontent and populism. He advocates for place-sensitive policies rather than relying on urban-centered growth. From focus on that individuals will be hurt to regions getting affected as a whole. Alternative views stress cultural factors, market-driven efficiency, or mobility policies over territorial redistribution. He stress where the academics went wrong regarding non-urbanized areas → failed to acknowledge all externalities associated with urbanisation + considered territorial inequality as almost irrelevant + overestimated the capacity and willingness of individuals to move + overlooked the economic potential of lagging-behind and declining ares -​ Rodríguez-Pose, Lee, and Lipp (2021): argue that declining social capital, economic stagnation, and inequality in left-behind U.S. regions have driven support for populism, particularly Trump. They highlight how weak local networks and distrust fuel political dissatisfaction. Alternative views emphasize cultural identity, racial dynamics, or national economic factors over regional decline. The study uses statistical analysis, correlating social capital and economic decline with voting patterns. While methodologically robust, causality remains complex, requiring further research to isolate specific drivers of populism. Lecture 2 - Models of market integration and regional growth → beta and sigma convergence: -​ Barro and Sala-i-Martin (1991): argue that poorer U.S. regions grow faster than richer ones (β-convergence), though conditional on factors like human capital and institutions. Some challenge this, citing persistent disparities due to agglomeration effects or club convergence, where only similar economies catch up. The study uses econometric analysis on U.S. state income data, testing absolute and conditional convergence models. While robust, the findings assume stable long-term growth factors and may overlook institutional and technological shocks, yet provide strong empirical support for conditional convergence. They found low convergence rates (1-2% per year) -​ Williamson (1965): argues that regional inequality follows an inverted U-shape during development—rising with industrialization and later declining as growth spreads. Critics suggest disparities may persist due to agglomeration, institutional lock-in, or club  convergence, where only some regions catch up. Using cross-country data, Williamson identifies this pattern by correlating regional GDP with national income. While influential, the study’s reliance on aggregate data may oversimplify factors like institutions and globalization, though it remains a key foundation for regional development research. Solow model predicts beta convergence! Williamsons 4 factors that will initially increase regional inequality 1.​ Labor - migration 2.​ Capital (agglomeration) 3.​ Government investments / central government 4.​ Interregional linkages - linkages between regions when developed, more linked because of social change, technology Contrast Barro/Sala-i-Martin with Williamson: -​ Barro and Sala-i-Martin (1991) argue for β-convergence, where poorer regions grow faster, reducing disparities, though conditionally based on factors like human capital and institutions. In contrast, Williamson (1965) proposes an inverted U-shape (sigma-convergence), where regional inequality first rises with industrialization before declining as growth spreads. While Barro and Sala-i-Martin emphasize steady-state determinants and long-run convergence, Williamson highlights a dynamic process influenced by structural transformation. Williamson’s inverted U hypothesis predicts increased interregional income distribution gap with initial growth, then it diminishes. This is the contrast to Barro & Sala-i-Martin’s Solow-based hypothesis, who predicts a catch-up (shrinking gap) from initial growth by poorer regions.q The neoclassical Solow growth model predicts convergence because of diminishing capital returns→ negative relationship between growth and initial GDP Β-convergence (beta) = refers to the concept that poorer economies or regions grow faster than richer ones, leading to a reduction in income disparities over time. It suggests that economies move toward a common steady-state level of income. Absolute (unconditional) convergence = assumes that all regions or economies, regardless of their initial conditions (like institutions, human capital, or infrastructure), will eventually converge to the same level of income. Poorer regions grow faster than richer ones without any additional factors considered. Conditional convergence = argues that regions will only converge if they share similar structural factors, such as institutions, human capital, or savings rates. Growth rates differ based on these conditions, meaning poorer regions grow faster only when these factors are aligned. σ-convergence (sigma) = The formula for sigma convergence is = standard deviation of population / mean. It explains the variation between different regions. Coefficient of variation: the cross-sectional standard deviation divided by the mean → often referred to as the Sigma-convergence. Measures the standard deviation of the regions. Are the regions clustered close together or are they far apart? This can show us each year by year. [Standard deviation of population /mean] then you get the coefficient of variation, or you can take the log(of x). It measures the standard deviation of regions, are the regions clustered close together, or are they far apart? This can show us year by year, the standard deviation of population / mean = coefficient of divergence Lecture 3 - Historical evidence, convergence and the creation of factor markets → more about the sigma convergence: -​ Enflo and Rosés (2015): argue that Sweden's regional inequality (1860-2000) was driven by structural change, migration, and government policies. They suggest that state investments in infrastructure and education helped reduce disparities, although some inequality persisted. Alternatives emphasize market forces, migration, and technological change. The authors use quantitative data and historical analysis to explore regional income and migration trends. The conclusions are sound but could benefit from more  econometric analysis to strengthen causality, though the paper offers valuable insights into policy impacts. -​ Felice (2011): argues that Italy's regional disparities from 1891 to 2001 were driven by factors like industrialization and regional policies, with the North-South divide widening over time. Alternative views emphasize geographic, institutional, or global factors. The paper uses historical regional GDP data and statistical methods to analyze economic trends. The conclusions are sound, though more detailed econometric modeling could strengthen the analysis. The paper provides valuable insights into Italy's long-term regional development. Sweden's regional divide is from population density, and Italy have a divide in living standards (GDP per capita) Beta → measured with regression (negative correlation between the income and initial GDP per capita) Sigma → (measured by standard deviation or standard deviation divided by the mean), poor regions to the rich and how they are distributed. We want low sigma → high equality Decreasing sigma → convergence in income levels  -​ What theory brings the prediction about convergence? The Solow-Swan growth model predicts convergence, suggesting that poorer economies will grow faster than richer ones, reducing income disparities over time. -​ What is the difference between conditional and unconditional convergence? Unconditional convergence assumes all regions will eventually converge, while conditional convergence states that regions only converge if they share similar conditions (like institutions and capital). -​ What is the difference between σ- and β-convergence? σ-convergence measures the reduction in income inequality across regions over time. β-convergence measures how poorer regions grow faster than richer ones, aiming to reduce income disparities. -​ How do you measure unconditional and conditional convergence using σ- and β-convergence? β-convergence measures convergence by analyzing growth rates. σ-convergence measures the dispersion of income levels. Unconditional convergence is assessed with β-convergence, while conditional convergence is measured after controlling for differences in conditions. -​ Which is the measure preferred by Williamson? Why? Williamson prefers σ-convergence because it directly measures the reduction in income inequality, which is more observable across regions over time. Compare Sweden - Italy using Williamson framework with his 4 factors -​ Labor migration: Sweden 19th century, and in Italy it was labor migration during 1800-1910 -​ Capital agglomeration: Sweden had something called Rehn Meidnor, and in Italy they had industrial triangle in the interwar period, subsidies to the south post-war -​ Government policies: in Sweden they had “flytt bidrag” in the time during labor migration, in Italy -​ Interregional linkages: In Sweden they had improvement of transport and market integration Italy: Cassa per il Mezzogiorno 1952 -​ Big government initiative to help the south and to close the gap between north and south -​ Probably the largest regional scheme set up by a western European country in the period, at least in terms of the total amount of funds -​ Concentrated on the industrial sector -​ Convergence came to halt in the 1970s The Swedish model: Rehn-Meinder model -​ Help industry and not specific regions -​ Active labor market policy -​ Structural Change on the labor market, a lot of influence for the people in the north of Sweden -​ Unemployment soared in the North → the idea was to move people from the north -​ Peripheral regions could no longer compete with lower wages -​ This policy was canceled after a while Lecture 4 - The rise of cities -​ Glaeser and Resseger (2010): Urbanization goes hand in hand with growth! Urban economics suggests a strong link between city size and worker productivity, but this pattern only holds in highly educated cities in the U.S. In the least educated metropolitan areas, city size has little impact on productivity or income, while in the most educated areas, population size explains 45% of productivity variation. Some of the productivity effect is due to skilled workers choosing bigger cities, but this only explains part of the relationship. Two main theories of agglomeration: 1. Non-knowledge-based (e.g., transport cost reduction, capital availability, geographic advantages). 2. Knowledge-based (e.g., learning from others, innovation, and knowledge spillovers). The study explores whether urban density helps workers learn faster or accelerates innovation, affecting wage growth differently. Findings suggest that workers in skilled cities learn faster, but wage growth patterns do not fully confirm whether learning or innovation is the main driver. Conclusion: If human capital keeps rising, cities will become even more important for economic growth, unless technology replaces the need for physical proximity. -​ Remi and Vollrath (2015): Urbanization without growth! Some regions have urbanisation without any growth in GDP. Urbanization is often linked to economic development, but recent trends show that cities are growing even in poorer countries, a phenomenon called "urbanization without growth."  This study examines the historical relationship between urbanization and income levels from 1500 to 2010. Before the mid-20th century, urbanization mainly occurred in wealthy, rapidly developing countries. Since then, urbanization has been concentrated in poorer nations, shifting the location of mega-cities from rich to poor countries. The paper explores the causes of this shift, distinguishing between positive urbanization (driven by improved infrastructure) and negative urbanization (driven by policy biases and rural poverty), with the latter playing a greater role in recent trends. Conclusion: Urbanization without growth, observed in the late 20th century, is part of a long-term trend dating back to 1500. “How does he distinguish the 2 hypotheses with empirical tests?” -​ People living in cities have a better return on their education -​ Regressions showing what we expect someone with x years of experience have in wage -​ More experience → more wage (but then gradually flattens out) -​ Steeper wage profile in the cities (non metro areas vs metro areas) → more wage just being in the metro areas -​ “Move to the city to get more out of your degree” Summary articles to remember: Lecture 1 - introduction, why care about regions: Andres Rodriguez-Pose (2018) + Rodríguez-Pose, Lee, and Lipp (2021) Lecture 2 - models of market integration and regional growth: Barro and Sala-i-Martin (1991) + Williamson (1965) Lecture 3 - historical evidence, convergence and the creation of factor markets: Enflo and Rosés (2015) + Felice (2011) Lecture 4 - the rise of cities: Glaeser and Resseger (2010) + Remi and Vollrath (2015)