We study the sectoral reallocation of employment over time and across countries, with a focus on the rise of services. We document substantial changes in the ratio of aggregate employment to working-age population across countries that are not systematically related to productivity growth or income levels, yet tightly linked to the rise in services employment. We assess the quantitative contribution of changes in aggregate employment to the rise of services using an otherwise standard model of sectoral reallocation calibrated to time-series for the United States. The calibrated model implies a high elasticity of changes in aggregate employment to services: a one percentage point change in aggregate employment generates on average a 0.7 percentage point change in services employment. The implication is that actual changes in aggregate employment account for one-third of the rise in services, on average across countries, and up to one-half in countries with sustained employment increases. Institutional subscribers to the NBER working paper series, and residents of developing countries may download this paper without additional charge at www.nber.org.

We exploit substantial variation in land-market institutions across Indian states and detailed household-level panel data to assess the effect of land-market distortions on agricultural productivity. We develop a model of heterogeneous farms and distorted land markets, featuring (i) state-level barriers to land-market participation and (ii) idiosyncratic (farm-level) distortions to farm size. We separately identify and estimate the two sources of land-market distortions in each state. We find substantial differences across states in rental barriers with large negative effects on agricultural productivity. Distortions associated with land-market participation contribute substantially to agricultural productivity differences across Indian states. (JEL D24, O13, O18, Q12, Q15, Q24)

Developing countries are characterized by frictions that impede the mobility of workers across occupations and space. We disentangle the role of insecure property rights from other labor mobility frictions for the reallocation of labor from agriculture to non-agriculture and from rural to urban areas. We combine rich household and individual-level panel data from China and an equilibrium quantitative framework that features the sorting of workers across locations and occupations. We explicitly model the farming household and the endogenous decisions of who operates the family farm and who potentially migrates, capturing an additional channel of selection within the household. We find that land insecurity has substantial negative effects on agricultural productivity and structural change, raising the share of households operating farms by almost 30 percentage points and depressing agricultural productivity by more than 10 percent. Quantitatively, land insecurity is as important as all other labor mobility frictions. We measure a sharp reduction in overall labor mobility barriers over 2004-2018 in the Chinese economy, all of which can be accounted for by improved land security, consistent with reforms covering rural land in China during the period. JEL classification: O11, O14, O4, E02, Q1.

Our motivation for the "Symposium on Misallocation and Structural Transformation" is that the processes of resource allocation and structural change are, each individually and jointly, interwoven with the process of economic growth and development. The common thread that transpires these processes is the allocation of economy-wide inputs across production units (sectors, firms, farms, regions, tasks). There is a growing recognition that this allocation and how it interacts with input accumulation and within unit productivity growth is at the heart of economic growth. Understanding the mechanisms and underlying forces that lead to resource misallocation and structural change are crucial for interpreting how today's developed economies came to be, but particularly critical for today's lower income countries, for which growth and development remain elusive, and concrete policy guidance is paramount. A fundamental inquiry within the discipline of economics pertains to the determinants underlying why some countries are rich and others poor. The magnitude of the disparity in income per capita across nations is extremely large, a factor of more than 30-fold between the richest and poorest countries in the world (Jones 2016). The welfare implications associated with closing this income gap are staggering, which necessitates understanding the fundamental sources of these great disparities and the associated policy implications. A consensus in the literature has centred around the importance of labour productivity, and in particular total factor productivity (TFP), the effectiveness with which countries can turn given amounts of inputs such as capital and labour into output, in accounting for a substantial portion of the differences in income across nations (Klenow and Rodriguez-Clare 1997, Prescott 1998). Consequently, an essential follow-up question pertains to the fundamental drivers of differences in aggregate productivity across countries. A major area of research in macroeconomics over recent decades has revolved around the quantitative examination of the role for aggregate outcomes of resource allocation across heterogeneous production units within sectors (Restuccia and Rogerson 2008, Hsieh and Klenow 2009) and sectoral structural transformation (Gollin et al. 2002, Duarte and Restuccia 2010). These examinations are motivated by empirical findings illustrating wide differences among nations in the operational scale in production such as farm size in the agricultural sector or establishment size in the non-agricutural sector (Adamopoulos and Restuccia 2014, Bento and Restuccia 2017; 2021) and the disparities both in sectoral productivities and stages of structural transformation among nations (Caselli 2005, Restuccia et al. 2008, Duarte and Restuccia 2010). Considering production heterogeneity within sectors is motivated by the fact that in developed countries the reallocation of factors of production across production units explains a large chunk of productivity growth over time (Baily et al. 1992, Foster et al. 2008). If resources are misallocated across production units, aggregate productivity can be low even in situations when aggregate resources are constant. This analytical framework has proven invaluable, as it unveils instances where ostensibly homogenous macroeconomic environments across nations belie substantial heterogeneity in the effective returns or costs confronting producers, thereby exerting heterogeneous impacts on resource allocation patterns and aggregate outcomes (Hopenhayn 2014, Restuccia and Rogerson 2017). For instance, variations in regulatory frameworks and institutional and policy environments may engender disparate cost structures and market conditions for different producers, thereby influencing an allocation of resources that depresses productivity in the aggregate. The exploration of potential misallocations across production units within sectors has uncovered numerous instances wherein even well-intentioned policies or institutional frameworks generate substantial negative effects on aggregate productivity levels. A wide variety of policies and institutions in developing countries can distort factors of production across producers. Broadly speaking, the literature on misallocation has followed two approaches in quantifying its effects on aggregate productivity. The indirect approach uses a canonical model of heterogeneous firms and backs out the extent of misallocation from disparities in marginal products across producers, an approach popularized by the seminal work of Hsieh and Klenow (2009). This approach has revealed considerable degrees of misallocation in many different sectors and country contexts. The direct approach identifies specific policies, institutions, or frictions causing misallocation, measures them, and using structural models quantifies their implications. The research program under this approach has unveiled the role of labour market policies (Hopenhayn and Rogerson 1993), size dependent policies (Guner et al. 2008), credit market imperfections (Buera et al. 2011, Midrigan and Xu 2014), land reforms (Adamopoulos and Restuccia 2020), market power (Peters 2020), among others. See Restuccia and Rogerson (2013), Hopenhayn (2014) and Restuccia and Rogerson (2017) for recent reviews of the literature. The allocation of resources across broad sectors of the economy can also play an important role in understanding aggregate productivity. It is well documented, at least since the work of Kuznets (1957), that the process of development is accompanied by a process of structural change, whereby the composition of economic activity—measured as employment, value added, or consumption expenditure—shifts from agriculture, to manufacturing and then to services. A substantial amount of research in recent years has documented these patterns for today's more advanced economies over time and has developed macroeconomic models consistent with both the aggregate Kaldor facts and sectoral Kuznets-stylized facts (Herrendorf et al. 2014). The literature has focused on mechanisms generating structural change with income effects through non-homothetic preferences (Kongsamut et al. 2001, Echevarria 1997) and relative price effects through differences in technologies across sectors (Baumol 1967, Ngai et al. 2019, Acemoglu and Guerrieri 2008), or both (Boppart 2014, Comin et al. 2021). A standard formulation of non-homotheticities generating income effects of structural change are the Stone-Geary preferences, with a minimum requirement of food consumption, which imply that, when consumer income is low, a disproportionate amount is spent on food—even if relative prices of goods are constant. In a closed economy, these preferences imply that productivity in the agricultural sector is essential in understanding the prevalence of agricultural employment in low productivity countries and the movement of employment out of agriculture associated with agricultural productivity growth. A substantial amount of work documents that agricultural productivity is particularly low in developing countries and seeks to understand why this is, e.g. Restuccia et al. (2008), Adamopoulos et al. (2022). The relative price formulation generates shifts in the composition of economic activity from differences in technological progress or capital intensities across sectors. For example, considering the substitution between industry and services, if productivity growth in industry is faster than in services and the two goods are complementary in consumption, then there is reallocation of employment to services. In this setting, productivity growth in industry outpaces demand for industry goods leading to deindustrialization. A recent literature quantifies the role differences in sectoral productivity growth across countries in generating heterogeneous patterns of structural transformation and aggregate outcomes (Duarte and Restuccia 2010, Huneeus and Rogerson 2023, Nguyen 2024). A related literature studies why labour is slow in moving from rural to urban areas and from agriculture to non-agriculture, despite the large agricultural productivity gap in low income countries (Gollin et al. 2014). The agricultural productivity gap can reflect sectoral selection (Lagakos and Waugh 2013), or frictions that prevent the movement of labour out of agriculture, e.g., monetary cost and risk (Bryan et al. 2014), rural insurance networks (Munshi and Rosenzweig 2016), transportation costs (Asher and Novosad 2020), and land rights (Ngai et al. 2019, De Janvry et al. 2015). Recent work by Adamopoulos et al. (2024) shows that insecure land rights over farmland can be an important barrier to the movement of labour out of agriculture and into urban areas, and can have substantial agricultural and aggregate productivity implications when interacted with selection. An essential finding in the broad literature of structural transformation is the relevance of sectoral productivity in generating reallocation across sectors. As a result, there is a natural connection between the policies and institutions that generate misallocation across producers within a sector and hence aggregate productivity effects within a sector, and their impact on structural transformation. That is, the misallocation of resources within a sector can be an important source of heterogeneous paths of structural change, an issue that has predominantly been studied with a focus on the agriculture–non-agriculture split (Adamopoulos and Restuccia 2014). Understanding what the fundamental drivers of sectoral productivity, and as a result structural change, is critical for policy guidance. For example, restrictive land markets in less developed countries can depress agricultural productivity by misallocating land and other inputs across farms, constituting a relevant source of productivity that prevents the reallocation of labour out of agriculture and migration from rural to urban areas (Adamopoulos et al. 2022; 2024). Poor transport infrastructure can also be a source of low agricultural productivity by limiting spatial specialization and access to intermediate inputs, thus keeping the majority of the population in rural dispersed communities (Adamopoulos 2024). This symposium is comprised of a great set of papers in the areas of misallocation and structural transformation. While all papers have important implications for economic growth, resource allocation and structural change, narrowly speaking, the first three papers are on resource allocation, while the fourth is on structural transformation. A common methodological attribute of all these papers is the use of micro-level data to study macro-level issues. This is consistent with the recent trend in macro development to use a granular micro-to-macro approach to understand development from the ground up. The article by Castro and Sevcik ("Occupational choice, human capital, and financial constraints") considers an augmented neoclassical growth model with production heterogeneity to study the aggregate productivity effects of financial frictions. In their framework, credit constraints affect not only production decisions of entrepreneurs, who are restricted in their operational scale, but also dynamic investment decisions on human capital, which in turn affect the productivity of operating firms. In this setting, the misallocation of resources across firms induced by financial frictions depresses the returns to human capital investment, distorts occupational choices (misallocation of talent), and hence alters the firm-level productivity distribution in the economy. All these factors lead to a magnification of the aggregate productivity losses from financial frictions. Castro and Sevcik show that a calibrated version of the model can account for between one third to two thirds of the aggregate productivity gap between India and the United States and that the impact of financial frictions on human capital decisions is a quantitatively important source of the aggregate productivity gap. This article advances our understanding of productivity differences across countries by providing a plausible and quantitatively substantial mechanism linking institutional distortions, such as financial frictions that are more prevalent in less developed countries, to both physical and human capital accumulation, misallocation of resources and the observed productivity distribution that is affected by human capital investment. As a result, the article provides an important link between the forces of broad capital accumulation, misallocation of resources within a given set of producers and differences in producer-level productivity distribution—three essential areas of research linked together via differences in financial development across countries. The article by Lee and Shin ("The plant-level view of Korea's growth miracle and slowdown") analyzes the growth miracle of South Korea between 1967–2000 using micro (plant-level) data for the manufacturing sector. Korea is a relevant case of inquiry because its growth episode is one of the more outstanding experiences of convergence to leading industrialized countries in the post-World War II era. For instance, the growth in real GDP per capita between 1967 and 2000 is more than 13-fold, implying an annualized growth rate of more than 8%, which contrasts to the growth rate of leading countries of around 2% per year. This is a remarkable convergence episode that transformed the average income per person in Korea. An important source of the income convergence is the growth in labour productivity in the manufacturing sector, the focus of Lee and Shin's article. What factors are responsible for this miracle productivity experience? Learning about this experience may help understand policies and institutions that could be replicated elsewhere. Moreover, it represents an opportunity to assess standard facts for an individual country over time in its process of substantial economic development in contrast with the usual approach of facts involving observations across countries at different points in the development process. Lee and Shin's article focuses on analyzing the evolution of the plant size distribution, static allocative efficiency and business dynamism of the Korean manufacturing sector during its growth miracle (1967–2000) and the subsequent slowdown since 2000. They uncover some important and somewhat puzzling, surprising facts. First, the average plant size features an inverse-U pattern over time, with a peak in the late 1970s, whereas comparable data across countries suggest a positive relationship between average plant size and income per capita (Bento and Restuccia 2017). Second, efficiency gains (the inverse of allocative efficiency), a standard measure of misallocation in the literature (Hsieh and Klenow 2009), decreases modestly until 1983 but increases substantially afterwards. Third, there is no systematic correlation between the growth rate of manufacturing productivity and either the level or the change in average plant size or misallocation. However, business dynamism, measured by firm turnover (job creation and destruction), diminished substantially staring in 2000, coinciding with the decline in manufacturing productivity growth. Cerdeiro and Ruane ("China's declining business dynamism") study the evolution of business dynamism in China during the period between 2003 and 2018. During the sample period, China featured strong growth and substantial economic transformation. Using data for the manufacturing sector, the authors document five facts on business dynamism. First, there is a reduction in the share of output and inputs of young firms. Second, there is a reduction in life cycle growth of firms. Third, there is a decline in life-cycle growth of process efficiency/product quality and investment in intangibles. Fourth, younger firms have higher capital productivity than older firms, with the gap increasing over time. Fifth, the dispersion of capital growth and the responsiveness of capital growth to capital productivity have both declined. The authors consider a simple model of firm reallocation and growth to estimate that the lower life-cycle productivity growth of young firms reduced manufacturing productivity growth by 0.8 percentage points annually, and worsening allocative efficiency of capital between young and old firms reduced manufacturing TFP by 1.25% between the early 2000s and late 2010s. Finally, they document empirically that provinces with a larger percentage of state-owned enterprises feature lower business dynamism. The article by Cao, Chen, Xi, and Zuo ("Family migration and structural transformation") provides a contribution into the process of structural change, and in particular the reallocation of employment out of agriculture and into urban centres in the context of migration decisions by married couples. The migration from rural to urban centres is a prominent feature of economic development. The authors consider a multi-sector model of structural transformation with household decisions and spatial features. Using the economic context of China, where spatial reallocation is restricted to the availability of welfare services to registered households, they use detailed household- and individual-level data to estimate the gender barriers to migration of married couples and their effects on structural transformation, aggregate productivity and gender gaps. An important finding is that, qualitatively, the reduction in migration costs contributes substantially to structural transformation. The authors also find important gender differences in migration costs, with substantial effects on structural transformation, aggregate productivity and the gender income gap. Each of these papers contribute to a better understanding of the processes of resource allocation and structural change and help in parsing out an important set of underlying forces. Given the fundamental importance of resource allocation and structural transformation for growth and development, these areas of research, individually and jointly, are open for more work, particularly exploiting the recent methodological approach of combining micro and macro tools.

We examine empirically whether the level of data aggregation affects the assessment of misallocation in agriculture. Using data from Ugandan farmers, we document a substantial discrepancy between misallocation measures calculated at the plot and at the farm levels. Estimates of misallocation at the plot level are much higher than those obtained with the same data but aggregated at the farm level. Even after accounting for measurement error and unobserved heterogeneity, estimates of misallocation at the plot level are extremely high, with potential nationwide agricultural productivity gains of 562%. Furthermore, we find suggestive evidence that granular data may be more susceptible to measurement error in survey data and that data aggregation can attenuate the relative magnitude of measurement error in misallocation measures. Our findings suggest caution in generalizing insights on measurement error and misallocation from plot-level analysis to those at the farm level.

The Malawi Integrated Survey of Agriculture (Malawi-ISA) is part of a new generation of household surveys funded by the Bill & Melinda Gates Foundation (BMGF) and led by the Living Standards Measurement Study (LSMS) team in the Development Research Group (DECRG) of the World Bank to improve the quality and policy relevance of household-level data on agriculture in SubSaharan Africa. The Malawi ISA (National Statistical Office, 2012) incorporates an extended and comprehensive agricultural questionnaire on agricultural production and factor inputs, including land quality and rain.

Using detailed household-level data from Malawi on physical quantities of agricultural outputs and inputs, we measure farm total factor productivity (TFP), controlling for land quality, rain, and transitory shocks. We find that operated land size and capital are essentially unrelated to farm TFP, implying substantial factor misallocation. The agricultural output gain from a reallocation of factors to their efficient use among existing farmers is a factor between 1.7- and 2.8-fold. We provide suggestive evidence connecting misallocation with the extent of land markets and illustrate how an efficient allocation via rental markets can substantially reduce agricultural income inequality and poverty. (JEL D24, D31, I32, O13, O15, Q12, Q15)

The expansion in farm size is an important contributor to agricultural productivity in developed countries where more productive farms are larger, but in less developed economies the allocation of factor inputs to more productive farms is hindered. How do distortions to factor-input allocation affect farm dynamics and agricultural productivity? We develop a model of heterogeneous farms making cropping choices and investing in productivity improvements. We calibrate the model using detailed farm-level panel data from Vietnam, exploiting regional differences in agricultural institutions and outcomes. We focus on south Vietnam and quantify the effect of higher measured distortions in the North on farm choices and agricultural productivity. We find that the higher distortions in north Vietnam reduce agricultural productivity by 47%, accounting for 70% of the observed 2.5-fold difference between regions. Moreover, two-thirds of the productivity loss is driven by farms’ choice of lower productivity crops and reductions in productivity-enhancing investment, which more than doubles the productivity loss from static factor misallocation.Institutional subscribers to the NBER working paper series, and residents of developing countries may download this paper without additional charge at www.nber.org.

What accounts for income per capita and total factor productivity (TFP) differences across countries?We study resource misallocation across heterogeneous production units in a general equilibrium model where establishment productivity and size are affected by policy distortions.We solve the model in closed form and show that policy distortions have a substantial negative effect on establishment productivity growth, average establishment size, and aggregate productivity.Calibrating a distorted benchmark economy to U.S. data, we find that empirically reasonable variations in distortions generate reductions in aggregate TFP of more than 24 percent while slightly increasing concentration in the establishment size distribution.If distortions in addition lower the exit rate of incumbent establishments, as supported by some empirical evidence, the aggregate TFP loss doubles to 48 percent.

We use household-level panel data from China and a quantitative framework to document the extent and consequences of factor misallocation in agriculture. We find that there are substantial within-village frictions in both the land and capital markets linked to land institutions in rural China that disproportionately constrain the more productive farmers. These frictions reduce aggregate agricultural productivity by affecting two key margins: (1) the allocation of resources across farmers (misallocation) and (2) the allocation of workers across sectors, in particular the type of farmers who operate in agriculture (selection). Selection substantially amplifies the productivity effect of distortionary policies by affecting occupational choices that worsen average ability in agriculture.

We use household-level panel data from China and a quantitative framework to document the extent and consequences of factor misallocation in agriculture. We find that there are substantial within-village frictions in both the land and capital markets linked to land institutions in rural China that disproportionately constrain the more productive farmers. These frictions reduce aggregate agricultural productivity by affecting two key margins: (1) the allocation of resources across farmers (misallocation) and (2) the allocation of workers across sectors, in particular the type of farmers who operate in agriculture (selection). Selection substantially amplifies the productivity effect of distortionary policies by affecting occupational choices that worsen average ability in agriculture.

We quantify the role of geography and land quality for agricultural productivity differences across countries using high-resolution micro-geography data and a spatial accounting framework. The rich spatial data provide for each cell of land covering the entire globe, the potential yield for 18 crops, which measures the maximum attainable crop output given soil quality, climate conditions, terrain topography, and a given level of cultivation inputs. While there is considerable heterogeneity in land quality across space, even within narrow geographic regions, we find that low agricultural land productivity is not due to unfavourable geographic endowments. If countries produced current crops in each cell according to potential yields, the rich-poor agricultural yield gap would virtually disappear, from 214% to 5%. We also find evidence of additional aggregate productivity gains attainable through spatial reallocation and changes in crop production.

Gollin and Udry (2021) estimate the contribution of mismeasurement to productivity dispersion among production units and conclude that previous studies have overestimated the potential efficiency gains from factor reallocation. We show that this conclusion is incorrect based on their own empirical evidence, which instead corroborates the importance of misallocation emphasized in the macro-development literature. We also point out important limitations in the implementation of the plot-level analysis that overstates the importance of mismeasurement in understanding productivity differences.

We construct a new dataset for the average employment size of establishments across sectors and countries from hundreds of sources. Establishments are larger in manufacturing than in services, and in each sector they are larger in richer countries. The cross-country income elasticity of establishment size is remarkably similar across sectors, about 0.3. We discuss these facts in light of several prominent theories of development such as entry costs and misallocation. We then quantify the sectoral and aggregate impact of entry costs and misallocation in an otherwise standard two-sector model with endogenous firm entry, firm-level productivity, and sectoral employment shares. We find that observed measures of misallocation account for the entire range of establishment-size differences across sectors and countries and almost 50 percent of the difference in non-agricultural GDP per capita between rich and poor countries.

We exploit substantial variation in land-market institutions across Indian states and detailed household-level panel data to assess the effect of land-market distortions on agricultural productivity. We develop a model of heterogeneous farms and distorted land markets, featuring (a) state-level barriers to land-market participation and (b) idiosyncratic (farm-level) distortions to farm size. We use the framework to separately identify and estimate the two sources of land-market distortions in each state using farm data on productivity, land endowment, land-market participation, and operational farm size. We find substantial differences across states in land-rental barriers with large negative effects on agricultural productivity. An efficient reallocation of land in India increases agricultural productivity by 65 percent and by more than 100 percent in some states, with more than 50% of these effects attributed to state-level rental barriers. Distortions associated with land-market participation contribute substantially to agricultural productivity differences across Indian states.

We assess the effects of land markets on misallocation and productivity by exploiting policy-driven variation in land rentals across time and space arising from a large-scale land certification reform in Ethiopia, where land remains owned by the state. Our main finding from detailed micro panel data is that land rentals substantially reduce misallocation and increase agricultural productivity. Our evidence builds from an empirical difference-in-difference strategy, an instrumental variable approach, and a calibrated quantitative macroeconomic framework with heterogeneous household-farms that replicates, without targeting, the empirical effects. These effects are nonlinear, impacting more farms farther away from efficient operational scale, consistent with our theory. Using our model, we find that more active land markets reduce inequality, an important concern for the design of land policy. We also find that the positive effects of land markets are mainly driven by formal market rentals as opposed to informal rentals. Finally, our analysis also provides evidence that land markets increase the adoption of more advanced technologies such as the use of fertilizers.