Desenvolvimento Socioeconômico 2

STRUCTURAL CHANGE IN AN INTERDEPENDENT WORLD A GLOBAL VIEW OF MANUFACTURING DECLINE Kiminori Matsuyama Northwestern University Abstract This paper presents a simple model of the world economy in which productivity gains in manufacturing are responsible for the global trend of manufacturing decline and yet in a crosssection of countries faster productivity gains in manufacturing do not necessarily imply faster declines in manufacturing In doing so it aims to draw attention to the common pitfall of using the crosscountry evidence to test a closed economy model and argues for a global perspective in order to understand crosscountry patterns of structural change one needs a world economy model in which the interdependence across countries is explicitly spelled out JEL F43 O11 O14 O19 1 Introduction We live in the global economy where countries are interdependent with one another Most empirical studies of structural change however write down a closed economy model apply it to each country and use the crosscountry data to test the model Effectively they treat each country as an autarky as if these countries were still isolated fiefdoms in the Middle Ages or were located on different planets This paper presents a simple example demonstrating how misleading this com mon practice can be in the context of productivitybased theory of manufacturing employment decline1 In many countries manufacturing employment has been declining over time A common explanation for this general trend attributes it to productivity gains in manufacturing With the rise in productivity fewer workers are needed to produce Acknowledgments I thank Daron Acemoglu Nobu Kiyotaki Bob Lucas those who came to my practice talk at Chicago Fed several members of the audience at the EEA session as well as the referee for their feedback Email address Matsuyama kmatsuyamanorthwesternedu 1 Although the first draft of this paper was originally written in February 1998 and circulated under the title ProductivityBased Theory of Manufacturing Employment Decline A Global Perspective I did not continue working on it afterwards Only recently have I decided to resucitate it in view of the recent growing interest on the topic as evident from Acemoglu and Guerrieri 2008 Buera and Kaboski 2006 2007 Ngai and Pissarides 2007 and others Journal of the European Economic Association AprilMay 2009 723478486 2009 by the European Economic Association Matsuyama Structural Change in an Interdependent World 479 a higher volume of manufacturing goods Unless productivity gains also lead to an equally higher growth in demand for manufacturing goods some workers in the manufacturing sectors will have to switch jobs to satisfy the higher demand in other sectors such as services In crosssection of countries however it is difficult to find any evidence that a country with higher productivity growth in manufacturing experiences a faster decline in its manufacturing employment See for example Figure 47 of Obstfeld and Rogoff 1996 which shows that countries like Germany and Japan experience slower if any declines in their manufacturing employments than countries like the US and the UK And some countries in the Pacific Rim such as South Korea Hong Kong Taiwan Vietnam and Indonesia have not shown a decline in their manufacturing employment shares One might be tempted to read such crosscountry evidence as a rejection of the productivitybased theories of manufacturing employment decline2 Such a read ing however is unwarranted as it implicitly assumes that each country is a closed economy and offers an independent observation The productivitybased theory argues that when South Korea experiences high productivity growth in manufac turing manufacturing workers will have to move to other sectors somewhere in the world thereby accelerating the global trend of the manufacturing employment decline The productivitybased theory does not say that South Korean manufac turing workers would have to switch jobs they could be instead American British or Japanese workers Indeed to the extent that productivity gains in South Korean manufacturing cause a shift in its comparative advantage toward manufacturing we should expect the net effect on its national employment share to be ambiguous or even positive 2 A Ricardian Model of the World Economy The main message of this paper can be conveyed by many different models The following Ricardian model of the world economy is chosen mostly because of its simplicity The world consists of two economies Home and Foreign There are three goods the numéraire O the manufacturing good M and the services S The first two O and M can be traded costlessly whereas S is nontradeable There is no production of the numéraire good both economies are endowed with y units of the numéraire good The economies are also endowed with one unit of labor which can be converted to M or S by constant returns to scale technologies Let 2 For example after pointing out that Japans manufacturing employment share has been roughly level since the mid1970s Obstfeld and Rogoff 1996 p 225 argue that Considering that Japan has had exceptionally high productivity growth in manufacturing relative to services its experience is especially hard to square with productivitybased theories of manufacturing employment decline AMAM and ASAS denote the labor productivity in the two sectors As usual the asterisks denote Foreign variables To keep it simple let us assume that the two countries may differ only in labor productivity in M and S Let PM be the international price of M and WW and PSPS be the wage rate and the price of S at Home Foreign Perfect competition implies that PM WAM WAM PS WAS PS WAS if both economies produce M and S For the moment let us proceed under the assumption that this condition holds Later the necessary parameter restrictions will be imposed to ensure that this is indeed the case in equilibrium The Home representative consumer has the following form of StoneGeary preferences UCO CM CS CO γOα βMCM γθ βSCSθ1αθ for θ 1 θ 0 CO γOα CM γβM1αCSβS1α for θ 0 If γ 0 M has a smaller income elasticity of demand than S If θ 0 the direct partial elasticity of substitution between M and S σ 11 θ is less than one which means that an increase in the supply of M would cause a morethanproportionate decline in the relative price of M over S The Home representative consumer maximizes utility subject to the budget constraint CO PMCM PSCs y W This yields the Home demand schedules for O and S CO γO αy γO W γPM CS βSσ PSσ 1 αy γO W γPM βMσ PM1σ βSσ PS1σ Likewise the Foreign demand schedules for O and S are CO γO αy γO W γPM CS βSσ PSσ 1 αy γO W γPM βMσ PM1σ βSσ PS1σ The market clearing conditions for the tradeable O in the world economy and for the nontradeable S in each economy are given by CO CO 2y CS AS1 LM CS AS1 LM where LM and LM are the manufacturing share of employment at Home and Foreign Equations 14 can be combined to solve for the equilibrium value of the shares of manufacturing in employment as follows LM α2 1 AMAM γAM βMβSσ ASAM1σ 1 βMβSσ ASAM1σ LM α2 1 AMAM γAM βMβSσ ASAM1σ 1 βMβSσ ASAM1σ Both LM and LM must take a value strictly between zero and one in order for both economies to produce M and S This is ensured under the following parameter restrictions βMβSσ ASAM1σ α2 1 AMAM γAM 1 βMβSσ ASAM1σ α2 1 AMAM γAM 1 which are assumed to hold when conducting comparative static exercises REMARK Several readers have asked me why I have chosen a model with three goods two tradable goods and one nontradable good In particular what is the role of the numéraire good O in the analysis Just to show that productivity gains in the tradable sector M may cause labor to move from the tradable sector M to the nontradable sector S it would suffice to have a simpler model with two goods M and S However in order for such processes of structural change from the tradable M to the nontradable S to become interdependent across countries the presence of a second tradable sector is necessary Note that as can easily be seen in equation 5 productivity changes in M in one country can have effects on the other country only when α 0 The assumption that O is not produced with labor and instead endowed is not essential it is made solely to keep the algebra simple for a wide range of preference specification over M and S Note that equation 5 is independent of y and γO This simplicity is due to the two assumptions nonproduction of O and the CobbDouglas preferences between O and the other two goods M and S 3 Comparative Statics Structural Change in an Interdependent World To examine the effects of productivity gains in manufacturing let us focus on two cases that capture the productivitybased theory of manufacturing declines in its essentials The first case relies on incomeelasticity differentials across sectors M and S The second case relies on productivity differentials across sectors M and S 31 IncomeElasticity Differentials across M and S Suppose σ 1θ 0 and γ 0 Then equation 5 is simplified to LM 1 β α2 1 AMAM γAM β LM 1 β α2 1 AMAM γAM β where β βMβM βS Note that with σ 1θ 0 these expressions are independent of AS and AS It is easy to verify that ΔAMAM ΔAMAM 0 ΔLM 0 ΔLM 0 The condition γ 0 implies that M has a smaller income elasticity of demand than S which means that with constant prices an increase in the demand for M due to the higher income does not keep up with an increase in the supply The Matsuyama Structural Change in an Interdependent World 483 manufacturing employment thus declines because of the productivity gains in that sector7 This condition does not ensure however that a country with faster productivity gains in manufacturing experiences a larger decline in its manufacturing employment It is easy to verify that AMAM 0 AMAM sgn LM sgn α2 γAM LM 0 Note that an increase in AM affects the manufacturing employment at home LM through two distinctive channels The first is the income effect If γ 0 and hence the demand for M does not grow as fast as the national income the income effect implies that productivity growth at home leads to a decline in manufacturing employment at home The second is the trade effect To the extent that productivity gains at home are larger than abroad a shift in comparative advantage leads to an increase in manufacturing employment at home The combined effect of Home productivity growth in manufacturing on Home manufacturing employment is ambiguous Its effect on Foreign manufacturing employment is unambiguously negative because there is no income effect Hence if one regresses the national share of manufacturing in employment on the national manufacturing productivity growth in the crosscountry data the effect could be positive Such crosscountry evidence hence cannot be interpreted against the productivitybased theory of manufacturing employment decline 32 Productivity Growth Differentials across M and S Suppose now that γ 0 and σ 1θ 0 Then equation 5 becomes LM α2 1 AMAM βMβSσ ASAM1σ 1 βMβSσ ASAM1σ 7 LM α2 1 AMAM βMβSσ ASAM1σ 1 βMβSσ ASAM1σ 7 A large number of studies use nonhomothetic preferences to introduce incomeelasticity differentials across sectors as the main driving force behind structural change For the reference see Matsuyama 2008 Most of these studies use StoneGeary preferences to explain the decline of agriculture and the rise of manufacturing in the context of early industrialization The exceptions include Matsuyama 2002 and Foellmi and Zweimüller forthcoming which use hierarchical preferences to explain sequential birth and growth of new industries and Buera and Kaboski 2006 2007 which use hierarchical preferences to explain the decline of manufacturing and the rise of services 484 Journal of the European Economic Association It is easy to verify that AMAM AMAM ASAS ASAS 0 LM 0 LM 0 The condition σ 1θ 0 implies that M is not very substitutable with S As M becomes more productive S becomes scarcer An increase in the supply of M would thus bring down the relative price of M and drive up the relative price of S fast enough to shift labor away from the progressive M sector towards the stagnant S sector Through its relative supply effect the M sector experiences a decline in its employment because of its own productivity gains8 As in the previous case however the condition σ 1θ 0 does not ensure that a country whose manufacturing productivity grows faster relative to services experiences a larger decline in its manufacturing employment Again this is because an increase in AMAS affects the manufacturing employment at home LM through two distinctive channels The first is the relative supply effect which works to reduce the employment in the Home manufacturing The second is the trade effect which works in the opposite direction The combined effect of Home productivity growth in manufacturing on Home manufacturing employment is ambiguous Its effect on Foreign manufacturing employment is unambiguously negative as the first effect is absent there Again if one regresses the national share of manufacturing in employment on the national manufacturing productivity growth in the crosscountry data the effect could be positive which does not constitute any evidence against the productivitybased theory of manufacturing employment decline 4 Concluding Remarks Broadly stated the productivitybased theory of manufacturing employment declines argues that productivity gains in manufacturing are responsible for the global trend for manufacturing employment declines in time series observed in many countries For the purpose of delivering this message a closed economy model is quite adequate However it is wrong to test the predictions of such a closed economy model based on the crosscountry evidence under the false assumption that each country in the data were in autarky and offered an independent observation Such a closed economy model is not designed to explain crosscountry variations of manufacturing employment shares If one is interested in explaining crosscountry variations one needs to adopt a global perspective That is to say one must write down a world economy model where the interdependence among countries is explicitly spelled out 8 The classical contribution for this mechanism is Baumol 1967 This mechanism also plays an important role in Ngai and Pissarides 2007 and Acemoglu and Guerrieri 2008 Matsuyama Structural Change in an Interdependent World 485 Clearly this paper is not the first to point out the common pitfall of using the crosscountry evidence For example Matsuyama 1992 pointed it out in the context of reading the historical evidence of regional patterns of early industrial ization As many historians believe the Industrial Revolution might not have been possible without the Agricultural Revolution that had preceded it This statement is consistent with the evidence that countries and regions with less productive agricultural sectors were among the first to industrialize Acemoglu and Ventura 2002 built an endogenous growth model of the world economy which generates unbounded growth of the world economy and convergence of the growth rates across countries thereby demonstrating that crosscountry growth convergence should not be interpreted as the evidence against endogenous growth models see Ventura 1997 2005 for more on this issue In these models crosscountry implications differ from timeseries implications because countries interact with each other through international trade Once stated this point should be quite intuitive Yet in my view it has not attracted the attention it deserves And the vast majority of empirical studies in macroeconomic growth and development continue to test a closedeconomy model using the crosscountry data under the false assumption that each country offers an independent observation9 By drawing attention to the common pitfall that even good economists have stumbled over this paper aimed to highlight the need for a global perspective To guard against such pitfalls we need to keep reminding ourselves of the simple truth we live in an interdependent global economy and our planet the world economy is the only closed economy we know of10 References Acemoglu Daron and Veronica Guerrieri 2008 Capital Deepening and Nonbalanced Economic Growth Journal of Political Economy 116 467498 Acemoglu Daron and Jaume Ventura 2002 The World Income Distribution Quarterly Journal of Economics 117 659694 Baumol William J 1967 Macroeconomics of Unbalanced Growth The Anatomy of Urban Crisis American Economic Review 57 415426 9 This practice is common not only among macroeconomists but also among economists in other fields For example many researchers have examined the effectiveness of different national innova tion policies on encouraging the RD activities by looking at the crosscountry evidence In doing so they pay little if any attention to the possibility that the US or European patent policies affect the incentive for innovation for exporting firms throughout the world 10 One might expect that trade economists have already addressed these questions of how inter national trade affects the patterns of structural change or the effectiveness of national innovation policies Unfortunately and perhaps surprisingly to the outsider very little has been done in the trade literature either because they are primarily concerned with explaining the patterns and vol ume of trade flows taking the structural differences across countries as given What is needed is a greater collaboration and exchange of ideas between international trade and other fields 486 Journal of the European Economic Association Buera Francisco J and Joseph P Kaboski 2006 The Rise of the Service Economy Working paper Northwestern and Ohio State Universities BueraFranciscoJandJosephPKaboski2007ScaleandtheOriginsofStructuralChange Working paper Northwestern and Ohio State Universities Buera Francisco J and Joseph P Kaboski 2008 Can Traditional Theories of Structural Change Fit the Data Paper presented at the 2008 Congress of European Economic Association Milan Italy Foellmi Reto and Josef Zweimüller 2006 Income Distribution and DemandInduced Innovations Review of Economic Studies 73 941960 FoellmiRetoandJosefZweimüller2008IsInequalityHarmfulforInnovationandGrowth Price versus Market Size Effects Paper presented at the 2008 Congress of European Economic Association Milan Italy Foellmi Reto and Josef Zweimüller forthcoming Structural Change Engels Consumption Cycles and Kaldors Facts of Economic Growth Journal of Monetary Economics Matsuyama Kiminori 1992 Agricultural Productivity Comparative Advantage and Eco nomic Growth Journal of Economic Theory 58 317334 Matsuyama Kiminori 2000 A Ricardian Model with a Continuum of Goods under Nonho mothetic Preferences Demand Complementarities Income Distribution and NorthSouth Trade Journal of Political Economy 108 10931120 Matsuyama Kiminori 2002 The Rise of Mass Consumption Societies Journal of Political Economy 110 10351070 Matsuyama Kiminori 2008 Structural Change In The New Palgrave Dictionary of Eco nomics Second Edition edited by Steven N Durlauf and Lawrence E Blume Palgrave McMillan Ngai L Rachel and Christopher A Pissarides 2007 Structural Change in a MultiSector Model of Growth American Economic Review 97 429443 Obstfeld Maurice and Kenneth Rogoff 1996 Foundations of International Macroeconomics MIT Press Ventura Jaume 1997 Growth and Interdependence Quarterly Journal of Economics 112 5784 Ventura Jaume 2005 A Global View of Economic Growth In Handbook of Economic Growth Vol IB edited by Philippe Aghion and Steven N Durlauf Elsevier BV This article has been cited by 1 Francisco J Buera Joseph P Kaboski 2009 Can Traditional Theories of Structural Change Fit The DataCan Traditional Theories of Structural Change Fit The Data Journal of the European Economic Association 723 469477 J Econ Growth DOI 101007s1088701591223 Premature deindustrialization Dani Rodrik1 Springer ScienceBusiness Media New York 2015 Abstract I document a significant deindustrialization trend in recent decades that goes con siderably beyond the advanced postindustrial economies The humpshaped relationship between industrialization measured by employment or output shares and incomes has shifted downwards and moved closer to the origin This means countries are running out of industrialization opportunities sooner and at much lower levels of income compared to the experience of early industrializers Asian countries and manufactures exporters have been largely insulated from those trends while Latin American countries have been especially hard hit Advanced economies have lost considerable employment especially of the low skill type but they have done surprisingly well in terms of manufacturing output shares at constant prices While these trends are not very recent the evidence suggests both glob alization and laborsaving technological progress in manufacturing have been behind these developments The paper briefly considers some of the economic and political implications of these trends Keywords Deindustrialization Industrialization Economic growth JEL classification 014 1 Introduction Our modern world is in many ways the product of industrialization It was the industrial revolution that enabled sustained productivity growth in Europe and the United States for the first time resulting in the division of the world economy into rich and poor nations It was industrialization again that permitted catchup and convergence with the West by a relatively smaller number of nonWestern nationsJapan starting in the late Nineteenth century South Korea Taiwan and a few others after the 1960s In countries that still remain mired in poverty B Dani Rodrik danirodrikharvardedu 1 John F Kennedy School of Government Harvard University Cambridge MA 02138 USA 123 J Econ Growth such as those in subSaharan Africa and south Asia many observers and policy makers believe future economic hopes rest in important part on fostering new manufacturing industries Most of the advanced economies of the world have long moved into a new postindustrial phase of development These economies have been deindustrializing for decades a trend that is particularly noticeable when one looks at the employment share of manufacturing Employment deindustrialization has long been a concern in rich nations where it is associated in public discussions with the loss of good jobs rising inequality and a potential decline in innovation capacity1 In terms of output deindustrialization has been in fact less striking and uniform a pattern that is obscured by the frequent reliance on value added measures at current rather than constant prices In the United States manufacturing industries share of total employment has steadily fallen since the 1950s coming down from around a quarter of the workforce to less than a tenth today Meanwhile manufacturing valueadded MVA has remained a constant share of GDP at constant pricesa testament to differentially rapid labor productivity growth in this sector In Great Britain at the other end of the spectrum deindustrialization has been both more rapid and thorough Manufacturings share of employment has fallen from a third in the 1970s to slightly above 10 today while real MVA at 2005 prices has declined from around a quarter of GDP to less than 15 2 Across the developed world as a whole real manufacturing output has held its own rather well once we control for changes in income and population as I will show later in the paper The term deindustrialization is used today to refer to the experience mainly of these advanced economies In this paper I focus on a less noticed trend over the last three decades which is an even more striking and puzzling pattern of deindustrialization in low and middleincome countries With some exceptions confined largely to Asia developing coun tries have experienced falling manufacturing shares in both employment and real value added especially since the 1980s For the most part these countries had built up modest manufac turing industries during the 1950s and 1960s behind protective walls and under policies of import substitution These industries have been shrinking significantly since then The low income economies of SubSaharan Africa have been affected nearly as much by these trends as the middleincome economies of Latin Americathough there was less manufacturing to begin with in the former group of countries Manufacturing typically follows an inverted Ushaped path over the course of develop ment Even though such a pattern can be observed in developing countries as well the turning point arrives sooner and at much lower levels of income today In most of these countries manufacturing has begun to shrink or is on course for shrinking at levels of income that are a fraction of those at which the advanced economies started to deindustrialize3 Devel oping countries are turning into service economies without having gone through a proper experience of industrialization I call this premature deindustrialization4 There are two senses in which the shrinking of manufacturing in low and medium income economies can be viewed as premature The first purely descriptive sense is that these 1 The bulk of RD and patents originates from manufacturing In Europe for example close to twothirds of business RD spending is done in manufacturing even though the sector is responsible for only 1415 of employment and value added in aggregate Veugelers 2013 p 8 2 These numbers come from Timmer et al 2014 which is the principal data source I will use in the paper 3 See also Amirapu and Subramanian 2015 who document premature deindustrialization within Indian states 4 The term seems to have been first used by Dasgupta and Singh 2006 although Kaldor 1966 made much earlier reference to early deindustrialization in the British context I am grateful to Andre Nassif for the Kaldor reference 123 J Econ Growth economies are undergoing deindustrialization much earlier than the historical norms As I will show in Sect 6 late industrializers are unable to build as large manufacturing sectors and are starting to deindustrialize at considerably lower levels of income compared to early industrializers The second sense in which this is premature is that early deindustrialization may have detrimental effects on economic growth Manufacturing activities have some features that make them instrumental in the process of growth First manufacturing tends to be techno logically a dynamic sector In fact as demonstrated in Rodrik 2013 formal manufacturing sectors exhibit unconditional labor productivity convergence unlike the rest of the economy Second manufacturing has traditionally absorbed significant quantities of unskilled labor something that sets it apart from other highproductivity sectors such as mining or finance Third manufacturing is a tradable sector which implies that it does not face the demand constraints of a home market populated by lowincome consumers It can expand and absorb workers even is the rest of the economy remains technologically stagnant Taken together these features make manufacturing the quintessential escalator for developing economies Rodrik 2014 Hence early deindustrialization could well remove the main channel through which rapid growth has taken place in the past My focus in the present paper is on doc umenting deindustrialization trends against the background of these considerations rather than on examining their normative consequences I do spend some time to consider the underlying causes of these trends I present a simple theoretical framework in Sect 7 to help interpret the key empirical findings of the paper Two important differences across country groups in particular need explanation First advanced countries as a group have managed to avoid output deindustrialization unlike the bulk of developingcountriesSecondamongdevelopingcountriesAsiancountrieshaveexperienced no output or employment deindustrialization Note that these patterns refer to outcomes after income and demographic trends are controlled for I do not attempt here a fullfledged causal explanation for the patterns But the model is suggestive of the combinations of technol ogy and trade shocks that can account for the observed heterogeneity In brief productivity improvements appear to have played the major role in the advanced economies while glob alization features more prominently in accounting for industrializationdeindustrialization patterns within the developing world The conventional explanation for employment deindustrialization relies on differential rates of technological progress Lawrence and Edwards 2013 Typically manufacturing experiences more rapid productivity growth than the rest of the economy This results in a reduction in the share of the economys labor employed by manufacturing as long as the elasticity of substitution between manufacturing and other sectors is less than unity σ 1 As I show in Sect 7 however under the same assumptions the output share of manufacturing moves in the opposite direction To get both employment and output deindustrialization we need to make additional assumptions that the trade balance in manufactures becomes more negative or that there is a secular demand shift away from manufactures The math is worked out in Sect 7 Since the more pronounced story in the advanced countries is employment rather than output deindustrialization a technologybased story does reasonably well to account for the patterns there Further the evidence suggests that a particular type of technological progress of the unskilledlabor saving type is responsible for the bulk of the labor displacement from manufacturing Sect 5 For developing countries however it is less evident that the technology argument applies in quite the same way Crucially the mechanism referred to above relies on adjustments in domestic relative prices Differential technological progress in manufacturing depresses the 123 J Econ Growth relative price of manufacturing In the case where σ 1 this decline is sufficiently large that it ensures demand for labor in manufacturing is lower in the new equilibrium The big difference in developing countries is that they are small in world markets for manufactures where they are essentially price takers In the limit when relative prices are fully determined by global rather than domestic supplydemand conditions more rapid productivity growth in manufacturing at home actually produces industrialization not deindustrialization in terms of both employment and output as the model of Sect 7 shows So the culprit for deindustrialization in developing countries must be found elsewhere The obvious alternative is trade and globalization A plausible story would be the follow ing As developing countries opened up to trade their manufacturing sectors were hit by a double shock Those without a strong comparative advantage in manufacturing became net importers of manufacturing reversing a long process of importsubstitution5 In addition developing countries imported deindustrialization from the advanced countries because they became exposed to the relative price trends originating from advanced economies The decline in the relative price of manufacturing in the advanced countries put a squeeze on manufacturing everywhere including the countries that may not have experienced much technological progress This account is consistent with the strong reduction in both employ ment and output shares in developing countries especially those that do not specialize in manufactures It also helps account for the fact that Asian countries with a comparative advantage in manufactures have been spared the same trends In sum while technological progress is no doubt a large part of the story behind employ ment deindustrialization in the advanced countries in the developing countries trade and globalization likely played a comparatively bigger role The outline of the paper is as follows In Sect 2 I discuss the data various measures of deindustrialization and the inverseU shaped relationship between industrialization and incomes In Sects 3 and 4 I document the patterns of deindustrialization over time and across different country groups In Sect 5 I look at employment deindustrialization differentiating across different labor types In Sect 6 I make the concept of premature deindustrialization more precise In Sect 7 I develop an analytical framework within which the empirical results can be interpreted In Sect 8 I conclude 2 The inverse Ushaped curve in manufacturing data measures and trends I begin by providing some indicators of changes in global manufacturing activity in recent decades Table 1 The data come from the United Nations and have globally comprehensive coveragebuttheygobackonlyto1970Thetoppanelofthetableshowstheglobaldistribution of manufacturing output while the lower panel shows shares of manufacturing in GDP for major regions Two key conclusions stand out First there has been a significant relocation of manufacturing from the richer parts of the world United States and Europe to Asia particularly China Second the share of manufacturing in GDP has moved differently in various regions and not always in a manner that would have been expected a priori Some lowincome regions subSaharan Africa and Latin America have deindustrialized while some highincome regions namely the US have avoided that fate 5 This echoes the concern in the voluminous literature on the Dutch disease that developing countries with comparative advantage in primary products would experience a squeeze on manufacturing as they open up to trade See Corden 1984 van Wijnbergen 1984 and Sachs and Warner 1999 123 J Econ Growth Table 1 Indicators of global manufacturing activity in 2005 constant USD World United States Western Europe Latin America and Caribbean Asia excluding China China SubSaharan Africa Other Shares in global MVA 1970 100 026 024 006 015 000 001 027 1980 100 022 021 008 018 001 001 029 1990 100 021 019 006 024 002 001 026 2000 100 024 016 007 024 006 001 022 2010 100 020 013 006 026 016 001 018 2013 100 019 013 006 026 018 001 017 MVA share in GDP 1970 017 013 022 020 016 009 014 1980 016 012 020 020 016 016 015 1990 016 012 019 019 019 018 015 2000 017 013 018 019 019 029 013 2010 018 013 018 017 021 036 011 2013 018 013 018 016 020 036 011 Source Calculated from United Nations httpunstatsunorgunsdsnaamaselbasicFastasp The UN does not provide manufacturing data for China for the period before 2005 Manufacturing is grouped under the aggregate Mining Manufacturing Utilities ISIC CE I have imputed manufacturing values for China by backward extrapolation of the manufacturing share in this aggregate 123 J Econ Growth There are a variety of industrializationdeindustrialization measures in the literature Some studies focus on manufacturing employment as a share of total employment while others use manufacturing output MVA as a share of GDP MVA shares in turn can be calculated at constant or current prices Different measures yield different trends and results For complete ness I will use all three measures in this paper denoting them as manemp manufacturing employment share nommva MVA share at current prices and realmva MVA share at constant prices I will focus in later sections on the real magnitudes manemp and realmva as nommva conflates movements in quantities and prices which are best kept distinct when trying to understand patterns of structural change and their determinants My baseline results are based on data from the Groningen Growth and Development Center GGDC Timmer et al 2014 These data span the period between the late 1940searly 1950s through the early 2010s and cover 42 countries both developed and developing The major economies in Latin America Asia and subSaharan Africa are included alongside advanced economies For more details on the data set see the Appendix Constantprice series are at 2005 prices6 For robustness checks and further analysis I will supplement this data with two other sources The Socioeconomic Accounts of the World InputOutput Database Timmer 2012 provide a disaggregation of sectoral employment by three skill categories for 40 mainly advanced economies And researchers at the Asian Development Bank have recently put together manufacturing employment and output series for a much larger group of countries using a variety of sources including the ILO UN and World Bank though these data start from 1970 at the earliest Felipe and Rhee 20147 I will combine these various sources on manufacturing with income and population data from Maddison 2009 updated using the World Banks World Development Indicators The income figures are at 1990 international dollars Figure 1 shows the simulated relationship between the three measures of industrialization and income per capita The figure is based on a quadratic estimation using country fixed effects and controlling for population size and period dummies See Sect 3 for the exact specificationThecurvesaredrawnforarepresentativecountrywiththemedianpopulation in the sample 27 million Period and country effects are all averaged to obtain a typical relationship for the sample and full time span covered The estimation results underlying the figure are shown in Table 2 cols 13 The quadratic terms are statistically highly significant for all three manufacturing indicators The share of manufacturing tends to first rise and then fall over the course of development However the turning points differ significantly In particular manemp peaks much earlier than realmva The employment share of manufacturing starts to fall past an income level of around 6000 in 1990 US after having reached an estimated maximum close to 20 Manufacturing output at constant prices peaks very late in the development process The estimated income level at which it begins to fall is in fact higher than any of the incomes observed in the data set above 70000 in 1990 US8 As we shall see in Sect 6 post1990 6 The only exception is West Germany for which there are no data after 1991 and constantprice series are at 1991 prices Since all my regressions include country fixed effects this difference in base year will be absorbed into the fixed effect for the country 7 I am grateful to Jesus Felipe for making these data available to me 8 These differences are statistically significant The 95 confidence intervals for log incomes at which manu facturing shares peak computed using the delta method are as follows manemp 845 897 nommva 879 958 and realmva 1016 1227 The confidence interval for manemp and nommva does not overlap that for realmva The series for manemp and nommva easily pass the Lind and Mehlum 2010 test for the presence of an inverse Urelationship in log GDP per capita while realmva fails it because the extremum occurs outside the observed income range 123 J Econ Growth 0 005 01 015 02 025 03 6 62 64 66 68 7 72 74 76 78 8 82 84 86 88 9 92 94 96 98 10 102 104 106 108 11 112 114 116 118 12 manemp nommva realmva Fig 1 Simulated manufacturing shares as a function of income In GDP per capital in 1990 international dollars data indicate a much earlier decline at less than half the pre1990 income level Note that the peak shares themselves are less meaningful in the case of output as they depend on the base year selected for converting current prices to constant prices The literature focuses on two possible explanations for why manufacturings share eventu ally falls Ngai and Pissarides 2004 Buera and Joseph 2009 Foellmi and Zweimuller 2008 Lawrence and Edwards 2013 Nickell et al 2008 One is demandbased and relies on a shift in consumption preferences away from goods and towards services This on its own would not produce the timing difference in peaks as a pure demand shift would have similar effects on manufacturing quantities output and employment The second explanation is techno logical and relies on more rapid productivity growth in manufacturing than in the rest of the economy As long as the elasticity of substitution is less than one this produces a decline in the share of manufacturing employment but not in the share of manufacturing output We need a combination of supply and demandside reasons to explain both the decline in manufacturings share and the later turnaround in output compared to employment An added complication is that the effects of technology and demand shocks depend cru cially on whether the economy is open to trade or not Matsuyama 2009 For the moment I leave these questions aside I will develop the analytical results linking technology demand and trade to deindustrialization in Sect 7 As Fig 1 shows nommva also peaks much earlier than realmva though not so early as manemp The explanation for this difference has to do with relative price changes over the course of development The relative price of manufacturing tends to decline as countries get richer tending to depress the share of MVA at current prices Figure 2 displays the pattern for four of the countries in our sample The relative price of manufacturing has more than halved in the United States since the early 1960s Great Britain has experienced a somewhat smaller decline In South Korea which has grown extremely rapidly manufacturings relative price 123 J Econ Growth Table 2 Baseline regressions Common sample Largest sample manemp nommva realmva manemp nommva realmva ln population 0115 0021 0142 0029 0113 0028 0122 0021 0192 0027 0039 0025 ln population squared 0000 0001 0002 0001 0005 0001 0001 0001 0004 0001 0003 0001 ln GDP per capita 0321 0027 0230 0031 0204 0025 0316 0026 0266 0031 0262 0027 ln GDP per capita squared 0018 0002 0013 0002 0009 0001 0018 0002 0014 0002 0012 0002 1960s 0029 0005 0011 0006 0008 0005 0018 0004 0010 0006 0028 0007 1970s 0044 0006 0021 0007 0004 0006 0033 0005 0014 0007 0026 0008 1980s 0066 0007 0033 0008 0011 0007 0054 0006 0028 0008 0034 0009 1990s 0086 0009 0052 0009 0017 0008 0074 0008 0049 0009 0040 0010 2000s 0117 0010 0085 0010 0035 0009 0105 0009 0085 0010 0059 0011 Country fixed effects Yes Yes Yes Yes Yes Yes Number of countries 42 42 42 42 42 42 Number of observations 1995 1995 1995 2209 2128 2302 Robust standard errors are reported in parentheses Levels of statistitical signficance 99 95 90 123 J Econ Growth relative price deflator of manufacturing Year 1 1960 1970 1980 1990 2000 2010 2 3 4 USA GBR KOR MEX Fig 2 Relative price trends has come down by a whopping 250 In Mexico meanwhile relative prices have remained more or less flat These trends are also consistent broadly with a technologybased explanation for the manufacturing hump More rapid productivity growth in manufacturing reduces the relative price of manufactured goods through standard supplydemand channels This in turn causes nommva to reach an earlier peak than reamva as shown in Fig 1 3 Deindustrialization over time As Fig 1 makes clear deindustrialization is the common fate of countries that are growing My interest here is to check whether deindustrialization has been more rapid in recent periods For this purpose I use a basic specification that controls for the effect of demographic and income trends with quadratic terms for log population pop and GDP per capita y as well as country fixed effects Di The baseline regression looks as follows manshareit β0 β1 ln popit β2 ln popit2 β3 ln yit β4 ln yit2 Σi γi Di ΣT φT PERT ϵit where manshare denotes one of our three indicators of industrialization Country fixedeffects allow me to take into account any countryspecific features geography endowments history that create a difference in the baseline conditions for manufacturing industry across different nations My main focus is on trends over time which are captured using period dummies PERT for the 1960s 1970s 1980s 1990s and post2000 years The post2000 dummy covers the period 2000 through the final year in the sample 2012 The estimated coefficients on these dummies φT allow us to gauge the effects of common shocks felt by manufacturing in each of the time periods relative to the excluded pre1960 years Table 2 shows two versions of the baseline results for each of our three measures of manufacturing industry manemp nommva and reamva Columns 13 are restricted to a common sample so that the results are directly comparable across the measures Columns J Econ Growth 46 employ the largest sample possible The common samples have 1995 observations while the others range from 2128 to 2302 The results for manemp and nonmva are very similar across the two specifications In both cases we find a sizable and significant negative trend over time larger for manemp than for nonmva Using the estimates from the common sample the average country in our sample had a level of manemp that stood 117 percentage points lower after 2000 than in the 1950s and 88 percentage points 01170029 lower than in the 1960s The corresponding reductions for nommva are 85 and 74 points respectively The declines in realmva are smaller and in the common sample show up significantly only for the post1990 period Depending on whether we use the common or largest sample the post2000 negative shock is 3559 points relative to the pre1960 period Figure 3ac provide a visual sense of the results They plot the estimated coefficients for the period dummies along with a 95 confidence interval around them The figures show a steady decrease over time in manufacturing shares after controlling for income and demographic trends The decline is most dramatic for employment it is less pronounced but still evident after 1990 for real MVA Manufacturing employment and activity have gone missing in a big way The samples in Table 2 provide good coverage across developed and developing regions but the number of countries is limited to 42 To make sure that the results are representative of trends in other countries as well I turn to the ADB dataset which includes a much larger group of countries up to 87 for manemp and 124 for nommva and realmva The limitation in this case is that coverage begins in 1970 Felipe and Rhee 2014 So I include dummies for the 1980s 1990s and post2000 years only with the 1970s as the excluded period Note that the ADB data set provides two alternative series for MVA one using UN sources and the other using World Bank data The results are presented in Table 3 and are quite similar to the previous ones Once again the strongest downward trend over time is for manemp a reduction of 65 points compared to the 1970s This matches up well with the corresponding number of 73 points 01170044 from Table 2 The decline in nommva is 30 or 52 points over the 1970s depending on which series is used Finally the decline for realmva is 0924 points 4 Deindustrialization in differenty country groups We can obtain some insight about the causes of these trends by looking at deindustrializa tion patterns in different country groups separately This is done in Tables 4 5 and 6 for manemp nommva and realmva respectively In each table the baseline regression is run for the following groups a developed countries b Latin American countries c Asian countries d subSaharan African countries and e subSaharan African countries exclud ing Mauritius Note that since there are no data for the 1950s for subSaharan Africa the period dummies for that region start from the 1970s Also two of the countries our global sample Egypt and Morocco do not belong in any of these groups so have been excluded from the grouplevel estimations9 9 An alternative and more efficient form of estimation would be to introduce periodXgroup dummies in a single global regression However the results in Tables 46 suggest considerable heterogeneity in the estimated coefficients on population and income terms across groups So allowing these coefficients to vary seems worth the price of potentially reduced power Since the period dummies in the groupspecific regressions are estimated tightly for the most part the loss in efficiency does not appear to make much practical difference 123 J Econ Growth 014 012 01 008 006 004 002 0 1960s 1970s 1980s 1990s 2000 014 012 01 008 006 004 002 0 1960s 1970s 1980s 1990s 2000 a b c 014 012 01 008 006 004 002 0 1960s 1970s 1980s 1990s 2000 Fig 3 Estimated period coefficients with 95 confidence intervals a manemp b nommva c realmva 123 J Econ Growth Table 3 Alternative data sets manemp nommva realmva ADBILO ADBUN ADBWB ADBUN ADBWB ln population 0196 0043 0194 0026 0260 0031 0008 0019 0044 0029 ln population squared 0004 0001 0004 0001 0007 0001 0001 0001 0002 0001 ln GDP per capita 0693 0052 0238 0016 0146 0019 0060 0015 0057 0017 ln GDP per capita squared 0039 0003 0013 0001 0008 0001 0002 0001 0002 0001 1980s 0023 0002 0011 0002 0002 0002 0006 0001 0000 0002 1990s 0043 0004 0029 0002 0010 0003 0016 0002 0003 0003 2000s 0065 0005 0052 0003 0030 0004 0024 0003 0009 0003 Country fixed effects Yes Yes Yes Yes Yes Number of countries 87 124 119 124 112 Number of observations 1947 4378 3691 5070 3312 Robust standard errors are reported in parentheses Levels of statistitical signficance 99 95 90 123 J Econ Growth Table 4 Country groups manemp All countries Developed countries Latin America Asia SubSaharan Africa SubSaharan Africa excl Mauritius ln population 0122 0021 0652 0122 0191 0032 0789 0102 0199 0019 0178 0014 ln population squared 0001 0001 0017 0003 0003 0001 0025 0003 0005 0001 0004 0000 ln GDP per capita 0316 0026 1070 0088 0902 0071 0912 0071 0190 0024 0148 0018 ln GDP per capita squared 0018 0002 0057 0005 0052 0004 0051 0004 0014 0002 0011 0001 1960s 0018 0004 0004 0004 0027 0004 0003 0013 na na 1970s 0033 0005 0021 0006 0050 0006 0016 0016 0002 0004 0003 0003 1980s 0054 0006 0052 0007 0079 0008 0022 0019 0004 0007 0021 0005 1990s 0074 0008 0072 0009 0096 0010 0013 0022 0007 0012 0033 0007 2000s 0105 0009 0096 0010 0131 0012 0004 0026 0007 0014 0035 0008 Country fixed effects Yes Yes Yes Yes Yes Yes Number of countries 42 10 9 11 11 10 Number of observations 2209 575 545 519 524 481 Robust standard errors are reported in parentheses Levels of statistitical signficance 99 95 90 123 J Econ Growth Table 5 Country groups nommva All countries Developed countries Latin America Asia SubSaharan Africa SubSaharan Africa excl Mauritius ln population 0192 0027 0752 0309 0223 0046 1009 0081 0552 0049 0519 0045 ln population squared 0004 0001 0016 0008 0007 0001 0029 0002 0017 0001 0014 0001 ln GDP per capita 0266 0031 1024 0139 0308 0157 0877 0054 0047 0061 0027 0056 ln GDP per capita squared 0014 0002 0059 0008 0016 0009 0050 0003 0007 0005 0006 0004 1960s 0010 0006 0003 0007 0001 0008 0008 0007 na na 1970s 0014 0007 0035 0010 0006 0010 0032 0010 0030 0005 0017 0005 1980s 0028 0008 0054 0011 0002 0014 0036 0014 0029 0008 0008 0009 1990s 0049 0009 0062 0013 0010 0018 0033 0018 0010 0010 0050 0013 2000s 0085 0010 0079 0015 0039 0020 0032 0022 0004 0012 0079 0016 Country fixed effects Yes Yes Yes Yes Yes Yes Number of countries 42 10 9 11 11 10 Number of observations 2128 451 498 576 565 512 Robust standard errors are reported in parentheses Levels of statistitical signficance 99 95 90 123 J Econ Growth Table 6 Country groups realmva All countries Developed countries Latin America Asia SubSaharan Afirca SubSaharan Afirca excl Mauritius ln population 0039 0025 4564 0776 0263 0027 0251 0084 0062 0029 0053 0031 ln population squared 0003 0001 0113 0019 0004 0001 0011 0003 0001 0001 0000 0001 ln GDP per capita 0262 0027 0778 0129 0135 0059 0737 0040 0123 0025 0106 0024 ln GDP per capita squared 0012 0002 0036 0008 0006 0003 0038 0003 0009 0002 0008 0002 1960s 0028 0007 0021 0011 0011 0004 0011 0006 na na 1970s 0026 0008 0007 0015 0017 0006 0027 0010 0017 0005 0012 0004 1980s 0034 0009 0006 0018 0052 0007 0034 0013 0015 0006 0004 0006 1990s 0040 0010 0013 0023 0078 0008 0041 0017 0011 0009 0022 0008 2000s 0059 0011 0021 0027 0101 0010 0044 0020 0003 0011 0042 0010 Country fixed effects Yes Yes Yes Yes Yes Yes Number of countries 42 10 9 11 11 10 Number of observations 2302 592 556 577 530 487 Robust standard errors are reported in parentheses Levels of statistitical signficance 99 95 90 123 J Econ Growth The results point to important regional differences First even though developed countries have experienced big losses in manemp and nommva they have done surprisingly well in realmva The estimated coefficients for the period dummies for realmva are in fact positive but statistically insignificant for the developed countries in recent decades Table 6 This is to be compared with significant negative estimates for Latin America and Africa once Mau ritius is excluded To be clear this does not mean that the rich nations have not experienced reductions in real manufacturing output shares in GDP It simply means that their experience can be well explained by income and demographic trends with little unexplained output deindustrialization left to account for in recent decades The results for Asia are even more striking Asia is the only region for which recent period dummies are not negative for manemp once again if Mauritius is excluded from the sub Saharan sample And the estimates for realmva in recent periods are actually positive and statistically significant These results suggest that Asia has not only bucked the global trend in manufacturing employment it has managed to maintain stronger manufacturing performance than would be expected on the basis of its income and demography The region that has done the worst is Latin America which has the most negative recent period effects for manemp and realmva The effects for nommva are not as pronounced suggesting that relative prices have not moved there against manufacturing nearly as much as in other regions Finally the estimates for subSaharan Africa depend heavily on whether Mauritius a strong manufactures exporter is included in the sample or not Without Mauri tius in the sample subSaharan African countries emerge as large losers on all three measures of industrialization Their output deindustrialization in recent decades looks especially dra matic in light of the strong showing for realmva in the 1970s captured by a positive and significant coefficient for dum1970s in Table 6 Since subSaharan countries are still very poor and widely regarded as the next frontier of laborintensive exportoriented manufactur ing these are quite striking findings The results with respect to Asia and the difference that the inclusion of Mauritius makes to the African performance strongly suggests that these variations in outcomes are related to patterns of comparative advantage and in particular how well or poorly countries have done in global trade in manufactures To test this idea I divide our sample of countries into two groups a manufactures exporters and b nonmanufactures exporters I use two criteria to split the sample based on the composition of trade The first classifies countries as manufactures exporters if the share of manufactures in exports exceeds 75 the second if the share of manufactures in exports exceeds the corresponding share in imports The results shown in Table 7 support the comparativeadvantage hypothesis Regardless of the criterion used the employment loss in manufactures exporters is smaller Whereas the period effects for realmva are strongly negative and significant for manufactures non exporters they change sign and are occasionally significant for manufactures exporters Regressions using ADB data with broader country coverage produce very similar results Table 8 The period effects for manemp are not distinguishable between the two groups with this sample but the realmva results show even stronger asymmetries In sum the geographical patterns of deindustrialization seem closely linked to globaliza tion Our results apparently reflect the sizable shift in global manufacturing activity in recent decades towards East Asia and China in particular with both Latin America and subSaharan Africa among the developing regions as the losers see Table 1 Countries with a strong com parative advantage in manufactures have managed to avoid declines in real MVA shares and employment losses where they have occurred have been less severe Interestingly on the output side it appears that the brunt of globalization and the rise of Asian exporters has been borne by other developing countries rather than the advanced economies What is partic 123 J Econ Growth Table 7 Results by manufacturing specialization Nonmanufactures exporters Manufactures exporters Manufactured exports 75 Share of manufactured exports share of manufactured imports Manufactured exports 75 Share of manufactured exports share of manufactured imports manemp realmva manemp realmva manemp realmva manemp realmva ln population 0202 0025 0146 0031 0174 0028 0130 0031 0326 0035 0034 0034 0444 0025 0184 0033 ln population squared 0003 0001 0001 0001 0002 0001 0001 0001 0009 0001 0002 0001 0014 0001 0007 0001 ln GDP per capita 0172 0021 0314 0051 0161 0022 0314 0051 0704 0043 0645 0021 0772 0042 0627 0025 ln GDP per capita squared 0010 0001 0018 0003 0009 0001 0017 0003 0039 0003 0033 0001 0042 0003 0032 0002 1960s 0032 0004 0055 0011 0028 0004 0057 0011 0004 0006 0004 0003 0002 0005 0007 0004 1970s 0057 0005 0070 0013 0054 0005 0073 0013 0004 0008 0024 0005 0002 0008 0030 0005 1980s 0080 0006 0087 0015 0078 0006 0091 0015 0025 0009 0014 0007 0020 0009 0022 0007 1990s 0093 0007 0097 0016 0093 0008 0101 0017 0057 0011 0013 0008 0050 0011 0019 0009 2000s 0120 0009 0123 0018 0123 0009 0128 0019 0089 0013 0012 0010 0079 0013 0014 0011 Country fixed effects Yes Yes Yes Yes Yes Yes Yes Yes Number of countries 26 26 26 26 16 16 16 16 Number of observations 1366 1426 1378 1425 843 876 831 877 Robust standard errors are reported in parentheses Levels of statistitical signficance 99 95 90 123 J Econ Growth Table 8 Results by manufacturing specialization ADBILOWB data Nonmanufactures exporters Manufactures exporters Manufactured exports 75 Share of manufactured exports share of manufactured imports Manufactured exports 75 Share of manufactured exports share of manufactured imports manemp realmva manemp realmva manemp realmva manemp realmva ln population 0130 0049 0094 0040 0130 0046 0131 0050 0036 0130 0132 0053 0022 01135 0116 0043 ln population squared 0002 0002 0004 0001 0002 0002 0006 0002 0001 0004 0005 0002 0001 0003 0004 0001 ln GDP per capita 0525 0056 0065 0017 0528 0061 0078 0018 0825 0008 0173 0032 0817 0069 0102 0031 ln GDP per capita squared 0030 0003 0003 0001 0030 0003 0004 0001 0045 0 005 0008 0 002 0045 0004 0003 0002 1980s 0028 0003 0008 0002 0028 0003 0007 0003 0018 0003 0018 0003 0019 0003 0010 0003 1990s 0042 0004 0016 0003 0042 0004 0013 0003 0049 0005 0023 0005 0049 0004 0009 0004 2000s 0066 0006 0028 0004 0066 0006 0026 00049 0069 0007 0034 0006 0071 0006 0015 0006 Country fixed effects Yes Yes Yes Yes Yes Yes Yes Yes Number of countries 55 80 52 73 32 32 35 39 Number of observations 1058 2411 1028 2238 889 901 919 1074 Robust standard errors are reported in parentheses Levels of statistitical signficance 99 95 90 123 J Econ Growth ularly striking is the magnitude of adverse employment effects in Latin America which is even larger than in developed economies 5 Employment deindustrialization by skill groups As the results above make clear deindustrialization shows up most clearly and in its strongest form in employment The only countries that have managed to avoid a steady decline in manufacturing employment in recent decades as a share of total employment are those with a strong comparative advantage in manufacturing The Socio Economic Accounts of the World InputOutput Database WIOD Timmer 2012 allow us to dig a bit deeper on the employment impacts These data provide a breakdown of manufacturing employment by three worker types lowskill mediumskill and highskill The data span the years 1995 2009 and include 40 countries with the coverage biased heavily towards Europe For the list of countries included see the Appendix I run essentially the same regression as before with two differences First the dependent variable is manufacturings share of the economys total employment of workers of a par ticular skill type Second since the data start from 1995 I use annual dummies rather than decade dummies As before there is a full set of country fixed effects This gives us three regressions one for each skill type Figure 4 plots the estimated coefficients for the year dummies The results are quite striking in that virtually the entire reduction over time in employment comes in the lowskill category Manufacturings share of lowskill employment has come down by 4 percentage points between 1995 and 2009 a decline that is statistically highly significant The decline in mediumskill employment is miniscule by comparison while manufacturings share of high 0045 004 0035 003 0025 002 0015 001 0005 0 0005 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 lowskill employment intermediateskill employment highskill employment Fig 4 Estimated year coefficients for employment of different skill types 123 J Econ Growth ITA 1980 GBR 1961 USA 1953 NLD 1964 SWE 1961 DNK 1962 JPN 1969 FRA 1974 KOR 1989 ESP 1975 GHA 1978 MUS 1990 NGA 1982 ZAF 1981 ZMB 1985 ARG 1958 BRA 1986 CHL 1954 COL 1970 CRI 1992 IDN 2001 IND 2002 MEX 1980 MYS 1997 PER 1971 0 1 2 3 4 6 7 8 9 10 peak manufacturing employment share Fitted values Fig 5 Income at which manufacturing employment peaks logs skill employment has actually slightly increased over the same period The chart underscores in a dramatic fashion that it is lowskill workers who have borne the lions share of the impact of recent changes in trade and technology on manufacturing 6 Premature deindustrialization Our results so far suggest that late industrializers will reach peak levels of industrialization as measured by manemp and realmva that are quite a bit lower than those experienced by early industrializers Let us denote these peak levels by manemp and realmva There is evidence that suggests these peak levels are reached at lower levels of income as well Denote that level of income by y Our baseline regressions capture the downward shift in the manufacturing hump over time but not the possibility that the curve may be moving closer to the origin as well Figure 5 suggests that manemp and y are in fact both lower for more recent indus trializers The figure displays manemp and y levels for the years of peak employment industrialization I have determined the turnaround years by looking at each country indi vidually and identifying visually the year at which manemp begins to decline Compare the two sets of countries at the opposite ends of the chart Industrialization peaked in Western European countries such as Britain Sweden and Italy at income levels of around 14000 in 1990 dollars India and many subSaharan African countries appear to have reached their peak manufacturing employment shares at income levels of 70010 Figure 5 represents a heuristic exercise that does not permit statistical testing To check more systematically how the industrialization inverse Ucurve has shifted over time I run 10 For a similar chart see Felipe and Rhee 2014 123 J Econ Growth Table 9 Regressions with interaction terms for post1990 manemp realmava ln population 0166 0019 0016 0025 ln population squared 0005 0001 0001 0001 ln GDP per capita 0326 0018 0273 0029 ln GDP per capita squared 0018 0001 0013 0002 ln GDP per capita X post1990 0031 0002 0015 0002 ln GDP per capita squared X post1990 0004 0000 0002 0000 Country fixed effects Yes Yes Number of countries 42 42 number of observations 2209 2302 Robust standard errors are reported in parentheses Levels of statistitical signficance 99 95 90 regressions that drop the period dummies and interact the income and income squared terms with a dummy for the post1990 period Using the 1990 year as a breakpoint is somewhat arbitrary But it ensures a sufficient number of observations on either side and is also useful as a demarcation of the period in which globalization gathered speed The results are shown in Table 9 The estimated coefficients on both interaction terms are statistically highly significant for manemp and realmva alike Moreover the signs confirm the pattern noted in Fig 5 Figures 6 and 7 plot the simulated industrialization levels against income for pre and post1990 based on the estimates in Table 9 We can see how the humpshaped curves have moved closer to the origin in the latter period in a particularly noticeably way for employment11 Using the same results we can calculate manemp and realmva and the corresponding point estimates for y for each subperiod These are displayed in Table 10 and show dramatic differences The table also shows 95 confidence intervals around the estimated y computed using the delta method The bands do not overlap in the case of manemp indicating that the pre and post1990 difference is clearly statistically significant The confidence intervals for realmva are much wider so they prevent us from reaching as strong a conclusion for output To summarize since 1990 countries have reached peak manufacturing employment and output shares at incomes that are around forty percent of the levels experienced before 1990 The employment effects are statistically highly significant The output effects which are almost equally large on average are also quite heterogeneous across different country groups So it is harder to reach a wholesale conclusion for realmva that pre and post1990 trends are statistically distinguishable 7 Some analytics To see how demand technology and trade shape the size of the manufacturing sector I consider a simple twosector model The determinants of the inverseU shaped pattern in manufacturing has been examined in the literature cited previously My focus here is not on this per se but on the forces behind the downward shift in manufacturing shares over time as 11 Amirapu and Subramanian 2015 present similar charts using industrial employment data from the World Development Indicators 123 J Econ Growth 0 005 01 015 02 025 6 62 64 66 68 7 72 74 76 78 8 82 84 86 88 9 92 94 96 98 10 102 104 106 108 11 112 114 116 118 12 pre1990 post1990 ln 1990 US Fig 6 Simulated manufacturing employment shares 005 0 005 01 015 02 025 03 6 62 64 66 68 7 72 74 76 78 8 82 84 86 88 9 92 94 96 98 10 102 104 106 108 11 112 114 116 118 12 pre1990 post1990 ln 1990 US Fig 7 Simulated manufacturing output shares MVAGDP at constant prices 123 Table 10 Maximum industrialization levels pre and post1990 manemp realmva Pre1990 Post1990 Pre1990 Post1990 Maximum share 215 189 279 241 Reached at income level GDP per capita in 1990 international 11048 4273 47099 20537 95 confidence interval 8785 14017 3831 4735 19667 112081 12429 34061 Source Authors calculations see text sectors This gives us our final equation which is the labormarket equilibrium equation βm pm θm αβm1 βn pn θn 1 α βn1 6 This equation equates the value marginal product of labor in the two sectors Since we can only determine relative prices lets take the nonmanufactured good to be the numeraire fixing pn at unity We are left with seven endogenous variables α qdn qsn qdm qsm pm and x We would need an additional global marketclearing equation to determine pm and x simultaneously This in turn requires modeling the rest of the world as well Here I will take a shortcut and make one of two extreme assumptions In one case prices are determined endogenously by developments in the home economy and net trade flows are exogenous In the second case the economy is sufficiently small that it remains a price taker in world markets so that x is endogenous and pm is a parameter13 These two characterizations are meant to capture the situations in the large developed and the developing countries respectively Consider first the advanced economy case Doing the comparative statics for the employment share of manufacturing we get dα ψ σ λσ θˆ m σ 1σ θˆ n 1 σ dxd qm d 7 where ψ 1α 1 βm 1 1 α 1 βn 1σ λ α βm 1 1 α βn 1 0 and λ qsm qdmd A lower trade surplus or bigger trade deficit in manufacturing dx 0 results in a smaller employment share in manufacturing which is not surprising Note that a reduction in x is formally analogous to an adverse demand shock for manufactures such as a secular shift in demand towards services and other nonmanufactures In both cases the manufacturing sector shrinks The relationship between technological progress θˆ m θˆ n and α on the other hand depends critically on the size of the elasticity of substitution in demand between manufactures and nonmanufactures Suppose for the moment that net trade in manufactures is small so that λ 1 Then if demand is inelastic σ 1 α is decreasing in technological progress in manufactures θˆ m and increasing in technological progress in nonmanufactures θˆ n More rapid TFP growth in manufacturing which is the usual case results in employment deindustrialization Intuitively the technological progressinduced reduction in the relative price of manufacturing does not spur demand for manufactures sufficiently so that the net result is a squeeze in manufactures employment These results are reversed when demand is elastic σ 1 This is the same as the finding in Ngai and Pissarides 200414 13 I assume that manufactures are the only goods that are traded In reality many services are also traded and the share that crosses national borders has increased over time Still even though services dominate the domestic economy they amount to less than a quarter of global trade For measurement and other issues posed by trade in services see World Trade Organization WTO 2010 14 Baumol and Bowen 1965 and Baumol 1967 are the classic works that looked at the consequences of lower productivity growth in services relative to manufactures The effect of technological progress in manufacturing however is also mediated through λ the ratio of supply to demand in manufacturing This is something that has not been emphasized in the earlier literature which typically assumes a closed economy Consider the case where a country is a large net importer of manufactures λ 1 As can be seen from 7 as long as σ λ 0 the coefficient that multiplies θˆ m is positive This is possible even when σ 1 and demand for manufactures is inelastic So we have a reversal of the result that inelastic demand and rapid technological progress in manufacturing produce employment deindustrialization The intuition behind this is as follows The lower the share of domestic supply in total consumption the smaller the effect of TFP in domestic manufactures on relative prices When manufacturing experiences rapid productivity growth it experiences less decline in relative prices compared to a country where domestic supply is a large share of domestic consumption Consequently domestic output and employment are larger in equilibrium In the limit when technological progress has no effect on domestic relative prices manufacturing employment is always boosted by TFP growth in manufactures This is indeed the case in our other benchmark example a small open economy which takes its relative prices from world markets Before we turn to that case however let us also look at the output share of manufacturing and how it is affected by trade and technology Denote the real value added share of manufacturing realmva by αq αq qsm qsm qsn We can now relate outputdeindustrialization dαq to employment deindustrialization dα as follows dαq αq 1 αq θˆ m θˆ n 1α βm 1 1 α βn dα 8 This shows that when the main shock comes from trade or demand with θˆ m θˆ n 0 the two measures of industrialization always move in the same way However when employment deindustrialization is due to differential TFP growth in manufacturing θˆ m θˆ n 0 it is possible for the output share of manufacturing to move very little or even to increase To see this in greater detail consider the case where the economy does not trade at all so that λ 1 In this case the output share of manufacturing must in fact rise We can read this off readily from the demandside relationship 3 Differential productivity growth in manufacturing depresses the relative price of manufacturing and this implies qdm qsm qdn qsn and therefore dαq 0 Or substituting 7 into 8 and solving we get dαq αq 1 αq σσ1 σσ1 1 α βm αβn θˆ m θˆ n Since the term in curly brackets is positive when σ 1 αq must move in the same direction as differential productivity growth in manufacturing This establishes that in an economy where trade plays a small role rapid technological progress in manufacturing produces employment deindustrialization but not output deindustrialization Let us look now at the smallopen economy case For this case we treat price changes parametrically and take x to be endogenous The comparative statics yields dα 1α 1βm 11α 1βn1 pm θm θn Technological progress in manufacturing now has an unambiguously positive effect on α Technological progress in nonmanufacturing has an unambiguously negative effect And an increase in the relative price of manufacturing works just like technological progress in manufacturing Moreover the result does not depend on σ or its magnitude as trade has the effect of delinking the supply side of the economy from the demand side For the same reason adverse domestic demand shocks would not produce deindustrialization in the small open economy domestic producers can sell the surplus output on world markets As Matsuyama 2009 has previously emphasized the relationship between productivity growth and industrialization depends crucially on whether we treat prices to be determined domestically or in the global economy This last set of results is important in interpreting the experience of developing countries that have experienced rapid deindustrialization These countries tend to be small in global markets for manufacturing so we can take treat them as price takers What equation 9 shows is that employment deindustrialization in those countries cannot have been the consequence of differentially rapid TFP growth in manufacturing at home That kind of technological progress would have fostered industrialization rather than the reverse In this respect developing countries are quite different from the advanced countries where there is considerable evidence that domestic technological progress was the culprit As price takers however these developing countries may have imported deindustri alization from the abroad Most countries in Latin America undertook significant trade liberalizations in the second half of the 1980s and early 1990s transforming themselves into open economies Many countries in SubSaharan Africa experienced trade opening as well around the same time As 9 shows a decline in the relative price of manufactures pm 0 the result of technological progress elsewhere the rise of China domestic trade liberalization or all three would have had the same effect as technological regress at home in manufacturing Even with more rapid TFP growth in manufacturing compared to nonmanufacturing these countries would find themselves deindustrializing in employment terms Putting it differently employment industrialization in the developing world requires more than differentially rapid TFP growth in manufacturing It requires that the productivity growth differential between manufacturing and nonmanufacturing also exceed the decline in manufactures relative prices on world markets Our empirical results suggest that only very few developing countries managed this feat consistently The configuration of analytical results under different assumptions about economic closure and the nature of the shocks is summarized in the table below Matsuyama 2009 shows that crosscountry results have to be interpreted with caution when economies are globally integrated In particular faster productivity gains need not be correlated with more rapid decline in manufacturing across countries even if productivity change is globally responsible for manufacturings decline See also Uy et al 2013 which develops a model with productivity and trade cost shocks under various assumptions about demand and uses it to explain South Koreas pattern of structural change The authors find that nonhomothetic demand more rapid productivity growth in manufacturing and the decline in manufacturing trade costs do a good job of explaining structural change with the exception of the decline in manufacturing after 1990 Springer J Econ Growth Effects of trade technology and demand on different measures of industrialization A Closed economy with σ 1 Effect on Technology shock ˆθm ˆθn 0 Trade shock dx 0 Adverse domestic demand shock on manufacturing manemp dα realmva dαq B Small open economy Effect on Technology shock ˆθm ˆθn 0 External price shock ˆpm 0 Adverse domestic demand shock on manufacturing manemp dα 0 realmva dαq 0 These are all ceteris paribus results for the case where each country can be treated indi vidually We need to exercise care in interpreting them when technology or demand shocks occur in many countries simultaneously requiring us to explicitly endogenize trade vol umes and relative prices on world markets Consider the consequences of technological progress in manufacturing that takes place in both developed and developing countries As longastheglobalsupplyofmanufacturesexceedssupplyofnonmanufacturesatunchanged prices the consequence is that ˆpm 0 for all countries Those countries which have expe rienced less technological progress in manufacturing will see their manufacturing industries suffer declines in output even though productivity has increased When TFP growth in manufacturing is global only those countries with the more rapid TFP growth will avoid deindustrialization So for Latin America or Africa to experience industrialization as open economies they must have had TFP growth in manufacturing that was faster than in the rest of the world which evidently did not happen Finally we can use this framework also to interpret the consequences of resource rents and related Dutchdisease issues in open economies Many Latin American and African countries have experienced booming primary sectors as a consequence of resource discoveries and a rise in commodity price In the present model a resource boom would manifest itself as an increase in productivity growth andor prices in nonmanufacturing relative to manufacturing in both cases The effects for small open economies can be read off the first two columns in part B of the table above the economy suffers both employment and output deindustrialization Effectively resource booms magnify the deindustrializing consequences that trade has on countries with comparative advantage in primary products16 8 Concluding remarks implications and consequences One way to understand the transformation in global manufacturing documented here is to consider the analogy of a closed economy with three regions a highincome region that is already industrialized and two lowincome regions one with a strong comparative advantage 16 Aid inflows operate similarly to resource booms in so far as they drive up the price of nontraded goods and reduce the relative price of manufactures For an examination of these issues in the SubSaharan African and Latin American contexts respectively see Rajan and Subramanian 2011 and Palma 2014 123 J Econ Growth or head start in manufacturing and the other without17 The economy experiences two shocks lowskill laborsaving technological progress and decline in transporttransaction costs among regions We would then observe the following patterns i a sharp decline in manufacturing employment in the highincome region with the impact on manufacturing output at constant prices depending on the balance between technology positive and trade negative shocks ii an increase in output and possibly employment in the lowincome region with the comparative advantage or head start in manufactures and iii a decline in both output and employment in the other lowincome region These consequences broadly capture the trends we have seen in the advanced economies Asian manufactures exporters and other developing economies respectively How concerned should low and middleincome countries be about their premature de industrialization The previous section treated productivity growth as exogenous to examine its consequences for industrialization under different economic structures But there is a reverse channel of causation as well especially in developing countries which goes from industrialization to economywide productivity In lowincome settings the movement of workers from the countryside to urban factories where their productivity tends to be much higher is an important source of productivity growth Industrialization contributes to growth both because of this reallocation effect and because manufacturing tends to experience rel atively stronger productivity growth over the medium to longer term In fact organized formal manufacturing appears to exhibit unconditional convergence which makes it a potent engine of growth Rodrik 2013 201418 From this perspective premature deindustrializa tion is not good news for developing nations It blocks off the main avenue of rapid economic convergence in lowincome settings19 The consequences are already visible in the developing world In Latin America as man ufacturing has shrunk informality has grown and economywide productivity has suffered In Africa urban migrants are crowding into petty services instead of manufacturing and despite growing Chinese investment there are as yet few signs of a significant resurgence in industry McMillan and Rodrik 2011 analyze employment patterns in a broad crosssection 17 I am grateful to an anonymous referee who suggested the closedeconomy analogy 18 Young November 2013 and forthcoming raises some important questions about gaps in intersectoral productivity and in manufacturing versus nonmanufacturing productivity growth arguing that these gaps may be due to selection based on unobserved worker skills To the extent that manufacturing is more productive because it employs the more capable workers it loses its specialness In particular more labor absorption in manufacturing would not raise economywide productivity as marginal workers drawn into manufacturing would be of the lowerproductivity type Even if selection effects are present however it is not clear they can explain why manufacturing industries that are further away from the frontier experience more rapid labor productivity growth as in Rodrik 2013 A recent paper by Franck and Galor 2015 suggests early industrialization may have adverse longrun effects within a country these authors find that regions in France that adopted industrial technology earlier eventually ended up with lower incomes and human capital levels It is not clear what the implications of such results to growth patterns across countries are however France is a postindustrial country and the crossregion findings are conditional on industrialization and withincountry convergence having taken place 19 A full welfare evaluation of the trends discussed in this paper must take into account other effects in addition to the foregone productivity gains due to premature deindustrialization For developing countries that are net importers of manufactures the global reduction in the relative price of manufactures due to technological progress in advanced countries represents a termsoftrade benefit and a static welfare gain For developing countries that are net exporters of manufactures there is a corresponding termsoftrade loss The fall in manufacturing prices may also reduce the cost of capitalgoods in developing countries and thereby spur investment Where private investment is suboptimal due to creditmarket or other failures this would represent an additional source of welfare gain It is in principle possible to attach some quantitative magnitudes for representative countries to each one of these effects using the results here and in Rodrik 2013 However such an effort would take this paper too far afield and I leave it to future work 123 J Econ Growth of developing economies and find that labor has been moving in the wrong direction in Latin America and Africa to low productivity services and away from high productivity activities such as manufacturing So structural change has ended up being growthreducing in these countries in recent decades unlike the 195075 period during which it made a strong positive contribution to growth see also de Vries et al 2013 and McMillan et al 201420 There has been no lack of growth in the developing world since the mid1990s But outside of Asia and the small group of manufactures exporters the evidence cited in the previous paragraph shows that this growth has not been driven by the traditional mechanism of industrialization Many of the growth booms appear to have been driven by capital inflows external transfers or commodity booms raising questions about their sustainability21 In the absence of sizable manufacturing industries these economies will need to dis cover new growth models One possibility is servicesled growth Many services such as IT and finance are high productivity and tradable and could play the escalator role that man ufacturing has traditionally played However these service industries are typically highly skillintensive and do not have the capacity to absorbas manufacturing didthe type of labor that low and middleincome economies have in abundance The bulk of other services suffer from two shortcomings Either they are technologically not very dynamic Or they are nontradable which means that their ability to expand rapidly is constrained by incomes and hence productivity in the rest of the economy None of this implies that developing countries have to stagnate As I discuss in Rodrik 2014 moderate growth is possible through improved fundamentalsbetter institutions and growing stocks of human capital skills and knowledge The advanced countries themselves have been able to grow at rates between 15 and 2 per annum despite declining manufactur ing But catching up with the frontier requires growth rates higher than these Sustained rapid convergence on the part of developing economies has historically required industrialization except for a very few resourcerich economies22 The political consequences of premature deindustrialization are more subtle but could be even more significant Historically industrialization played a foundational role in Western Europe and North America in creating modern states and democratic politics The labor movement a product of industrialization led demands for the expansion of the franchise and eventually the creation of the welfare state It was the bargain between elites and organized labor that enabled the development of democracy Acemoglu and Robinson 2009 The weakness of organized labor in todays developing societies is likely to foster different paths of political development not necessarily friendly to liberal democracy In particular the 20 It is possible that these trends will be reversed as manufacturing migrates from Asia to lowwage countries Anecdotal evidence eg the rise of Chinese manufacturing investment in countries such as Ethiopia and Rwanda as well as some of the more systematic evidence in McMillan et al 2014 suggest that manufacturing may have a renewal of sorts in SubSaharan Africa But the fact that we rarely see double humps in the manufacturing curve should make us skeptical of this eventuality 21 On Africa for example see African Center for Economic Transformation 2014 which emphasizes the need for productive diversification and structural transformation if recent growth rates are to be sustained 22 I show in Rodrik 2014 that the vast majority of the countries that experienced growth rates of 45 percent or more for at least three decades are those that underwent rapid industrialization The list is composed essentially of two categories of countries some in the European periphery during the 1950s and 1960s eg Spain Portugal Israel and some in East Asia since the 1960s eg South Korea Taiwan Malaysia The exceptions are some small but resourcerich countries eg Botswana Oman Equatorial Guinea many of which experienced reversals eventually In that paper I propose a framework that distinguishes between two channels of growth with overlapping but distinct requirements a fundamentals channel that relies on the accumulation of economywide skills and institutional capabilities and a structural transformation channel that relies on industrialization I argue that slowtomoderate growth is possible with the former but that rapid convergence requires the latter 123 J Econ Growth substitution of identity or ethnic cleavages for class cleavages as the central loci of politics may spawn electoral or illiberal democracies We present a formal model in Mukand and Rodrik 2015 which suggests precisely that Such considerations are necessarily speculative Nevertheless whatever the specific con sequences premature deindustrialization suggests the future of todays developing countries will be unlike the past of todays advanced societiesneither economically nor politically Acknowledgments I am grateful to Elias SanchezEppler Russell Morton and Juan Obach for expert research assistance Robert Lawrence and Arvind Subramanian for useful conversations David Romer for comments and Jesus Felipe for sharing his data set Four referees have provided very constructive suggestions that led to improvements Appendix Country and variable coverage in the GGDC 10Sector Database Acronym Country Value added in current prices Value added in constant prices Employment by sector SubSaharan Africa BWA Botswana 19642010 19642010 19642010 ETH Ethiopia 19612010 19612010 19612010 GHA Ghana 19602010 19602010 19602010 KEN Kenya 19602010 19642010 19692010 MWI Malawi 19602010 19662010 19662010 MUS Mauritius 19602010 19702010 19702010 NGA Nigeria 19602010 19602010 19602011 NGAalt Nigeria 2014 GDP revision 20102013 20102013 in 2010 prices SEN Senegal 19602010 19702010 19702010 ZAF South Africa 19602010 19602010 19602010 TZA Tanzania 19602010 19602010 19602010 ZMB Zambia 19602010 19652010 19652010 North Africa EGY Egypt 19602013 19602012 19602012 MOR Morocco 19702012 19602012 19602012 Asia CHN China 19522011 19522010 19522011 HKG Hong Kong 19702011 19742011 19742011 IND India 19502012 19502012 19602010 IDN Indonesia 19662012 19602012 19612012 JPN Japan 19532011 19532011 19532012 KOR South Korea 19532011 19532011 19632011 MYS Malaysia 19702011 19702011 19752011 123 J Econ Growth Acronym Country Value added in current prices Value added in constant prices Employment by sector PHL Philippines 19712012 19712012 19712012 SGP Singapore 19702012 19602012 19702011 TWN Taiwan 19512012 19612012 19632012 THA Thailand 19512011 19512011 19602011 Latin America ARG Argentina 19502011 19502011 19502011 BOL Bolivia 19582011 19502011 19502010 BRA Brazil 19902011 19502011 19502011 CHL Chile 19502011 19502011 19502012 COL Colombia 19502011 19502011 19502010 CRI Costa Rica 19502011 19502011 19502011 MEX Mexico 19502011 19502011 19502012 PER Peru 19502011 19502011 19602011 VEN Venezuela 19602012 19502012 19502011 North America USA United States of America 19472010 19472010 19502010 Europe DEW West Germany 19681991 19501991 1991 prices 19501991 DNK Denmark 19702011 19472009 19482011 ESP Spain 19702011 19472009 19502011 FRA France 19702011 19502009 19502011 GBR United Kingdom 19602011 19492009 19482011 ITA Italy 19702011 19512009 19512011 NLD The Netherlands 19702011 19492009 19502011 SWE Sweden 19702011 19502009 19502011 Source Timmer et al 2014 See httpwwwrugnlresearchggdcdata10sectordatabase Countries included in the Socio Economic Accounts of the World InputOutput Database WIOD Austria Germany Netherlands Canada China Belgium Greece Poland United States India Bulgaria Hungary Portugal Japan Cyprus Ireland Romania South Korea Czech Republic Italy Slovak Republic Australia Denmark Latvia Slovenia Brazil Taiwan Esto nia Lithuania Spain Mexico Turkey Finland Luxembourg Sweden Indonesia France Malta United Kingdom Russia Source Timmer 2012 latest update available at httpwwwwiodorgprotected3data updatesep12SEA20SourcesJune2014pdf 123 J Econ Growth References Acemoglu D Robinson J 2009 Economic origins of dictatorship and democracy Cambridge Cam bridge University Press African Center for Economic Transformation 2014 Africa Transformation Report 2014 Accra Ghana Amirapu A Subramanian A 2015 Manufacturing or services An indian illustration of a development dilemma Washington DC Center for Global Development and Peterson Institute Baumol W J 1967 Macroeconomics of unbalanced growth The anatomy of urban crisis The American Economic Review 573 415426 Baumol W J Bowen W G 1965 On the performing arts The anatomy of their economic problems The American Economic Review 5512 495502 Buera F J Kaboski J P 2009 Can traditional theories of structural change fit the data Journal of the European Economic Association 723 469477 Corden W M 1984 Booming sector and Dutch disease economics Survey and consolidation Oxford Economic Papers New Series 363 359380 Dasgupta S Singh A 2006 Manufacturing services and premature deindustrialization in developing countries A Kaldorian analysis UNUWIDER United Nations University Research Paper No 200649 de Vries K Timmer M P de Vries G J 2013 Structural transformation in Africa Static gains dynamic losses GGDC research memorandum GD136 Groningen Growth and Development Centre University of Groningen Felipe J Mehta A Rhee C 2014 Manufacturing matters but its the jobs that count Asian Develop ment Bank Economics Working Paper Series No 420 Foellmi R Zweimuller J 2008 Structural change Engels consumption cycles and Kaldors facts of economic growth Journal of Monetary Economics 55 13171328 Franck R Galor O 2015 Is industrialization conducive to longrun prosperity Brown University unpublished paper Kaldor N 1966 Causes of the slow rate of economic growth of the United Kingdom An inaugural lecture Cambridge England Cambridge University Press Lawrence R Z Edwards L 2013 US Employment deindustrialization Insights from history and the international experience Peterson Institute for International Economics Policy Brief No PB1327 Lind J T Mehlum H 2010 With or without u Theappropriate test for a Ushaped relationship Oxford Bulletin of Economics and Statistics 721 109118 Maddison A 2009 Statistics on World Population GDP and Per Capita GDP 12008 AD Available at httpwwwggdcnetmaddisonoriindexhtm Matsuyama K 1992 Agricultural productivity comparative advantage and economic growth Journal of Economic Theory 582 317334 Matsuyama K 2009 Structural change in an interdependent world A global view of manufacturing decline Journal of the European Economic Association 723 478486 McMillan M Rodrik D 2011 Globalization structural change and economic growth In M Bachetta M Jansen Eds Making globalization socially sustainable Geneva International Labor Organization and World Trade Organization McMillan M Rodrik D VerduzcoGallo Í 2014 Globalization structural change and productivity growth with an update on Africa World Development 63 1132 Mukand S Rodrik D 2015 The political economy of liberal democracy Unpublished paper Ngai L R Pissarides CA 2004 Structural change in a multisector model of growth Center for Economic Performance London School of Economics Nickell S Redding S Swaffield J 2008 The uneven pace of deindustrialization in the OECD Center for Economic Performance London School of Economics Palma J G 2014 Industrialization premature deindustrialization and the Dutch disease Revista NECAT 35 723 RodríguezClare A 1996 The division of labor and economic development Journal of Development Eco nomics 491 332 Rodrik D 1996 Coordination failures and government policy A model with applications to East Asia and Eastern Europe Journal of International Economics 4012 122 Rodrik D 2013 Unconditional convergence in manufacturing Quarterly Journal of Economics 1281 165204 Rodrik D 2014 The past present and future of economic growth In F Allen et al Towards a better global economy Policy Implications for Citizens Worldwide in the 21st Century Oxford Oxford University Press 123 J Econ Growth SachsJDWarnerAM1999ThebigpushnaturalresourceboomsandgrowthJournalofDevelopment Economics 591 4376 Timmer Marcel P de Vries GJ de Vries K 2014 Patterns ofStructural Change in Developing Countries Groningen Growth andDevelopment Center Research Memorandum 149 Timmer M P Ed 2014 The world inputoutput database WIOD Contents sources and methods WIOD Working Paper Number 10 Uy T Yi KM Zhang J 2013 Structural change in an open economy Journal of Monetary Economics 606 667682 van Wijnbergen S 1984 The dutch disease A disease after all The Economic Journal 94373 4155 VenablesAJ1996TradepolicycumulativecausationandindustrialdevelopmentJournalofDevelopment Economics 491 179197 Veugelers R Ed 2013 Manufacturing Europes future Brussels Bruegel World Trade Organization 2010 Measuring trade in services Geneva WTO Young A Structural transformation the mismeasurement of productivity growth and the cost disease of services American Economic Review forthcoming Young A 2013 Inequality the urbanrural gap and migration Quarterly Journal of Economics 128 1727 1785 123 1 UERJ Faculdade de Ciências Econômicas Disciplina Desenvolvimento Socioeconômico II Professor Thiago Sevilhano Martinez Período Segundo semestre de 2025 Lista de exercícios 2 Transformação Estrutural Entrega até domingo 14dezembro às 2350 As questões a seguir tratam dos determinantes da transformação estrutural a mudança de longo prazo na composição setorial das economias 1 Lei de Engel 25 pontos a 05 ponto Explique o que é a Lei de Engel do consumo b 1 ponto Use a Lei de Engel para interpretar o gráfico abaixo Fonte CMAP 2022 Desoneração de PISCOFINS sobre os produtos da Cesta Básica Disponível em Desoneração de PISCOFINS sobre os produtos da Cesta Básica Ministério do Planejamento e Orçamento Dados da Pesquisa Nacional de Orçamentos Familiares POF anos 20022003 20082009 e 20172018 c 1 ponto Como a Lei de Engel afeta o processo de transformação estrutural 2 Efeito Baumol 5 pontos 1 por item Seja uma economia fechada com dois setores de produção manufatura M e serviços S A função de utilidade da família representativa que habita esta economia tem formato CES e é dada por ucMcS βM cMθ βS cSθ1θ onde cM e cS são respectivamente o consumo de manufaturados e de serviços As famílias têm uma dotação total de horas de trabalho unitária LM LS 1 de oferta inelástica que podem ser usadas para trabalho na manufatura LM ou nos serviços LS O salário pago W por hora é igual nos dois setores Toda a renda y da família provém dos salários é gasta com o consumo dos dois bens O bem manufaturado tem preço pM e os serviços têm preço pS por unidade de consumo a Construa e resolva o problema de maximização de utilidade das famílias Mostre que vale a seguinte relação entre o consumo relativo e o preço relativo dos dois produtos cMcS βM pSβS pM11θ b A elasticidade de substituição entre manufaturados e serviços é definida por σ logcMcSlogpMpS A sigla CES significa constant elasticity of substitution Verifique que a elasticidade de substituição no consumo de manufaturados e serviços é de fato constante nesta economia Para isso a partir da relação demonstrada no item anterior mostre que a elasticidade de substituição será dada por σ 11θ c Reescreva a expressão do final do item a em termos de σ e substitua na restrição orçamentária do consumidor Faça manipulações adicionais para mostrar que pS cSy 11 βMβSσ pMpS1σ Se os manufaturados e serviços fossem bens substitutos com σ 1 o que aconteceria com a participação do gasto com serviços na renda caso houvesse um aumento do preço dos serviços pS em relação ao preço dos manufaturados pM E se manufaturados e serviços fossem complementares com σ1 d As firmas desta economia operam em concorrência perfeita com as seguintes funções de produção YM AM LM YS AS LS onde YM e YS são as quantidades produzidas LM e LS são os insumos de trabalho e AM e AS são a produtividade do trabalho nos setores de manufaturados e serviços respectivamente O custo salarial das firmas em ambos os setores é de W por hora trabalhada Construa e resolva o problema de maximização de lucros das firmas Mostre que o preço relativo é determinado pela produtividade relativa através de pMpS ASAM e Substitua a relação de preços relativos do item d na expressão dada no item c depois use as condições de market clearing para encontrar a expressão que determina a fração da mão de obra alocada nos serviços LS em função da produtividade relativa ASAM Usando essa expressão apresente o raciocínio de Baumol para explicar a queda da participação da indústria e aumento da participação dos serviços na alocação da força de trabalho de economias maduras Inclua na sua explicação o papel do valor da elasticidade de substituição σ 3 Abertura comercial 25 pontos Matsuyama 2009 apresenta um modelo para ilustrar como a abertura comercial faz com que as trajetórias de transformação estrutural dos países afetem umas às outras Há duas economias doméstica e estrangeira Neste modelo a utilidade das famílias é dada por ucO cM cS cO γOα βM cMθ βS cSθ1αθ onde cO é o consumo de um bem tradable extraído de recursos naturais que é o bem numerário pO 1 cM é o consumo de bens tradables manufaturados e cS o consumo de serviços nontradables Os parâmetros adicionais em relação ao modelo da questão anterior são 0 α 1 e γ0 0 Agora a restrição orçamentária do consumidor doméstico é cO pM cM pS cS y onde y é a renda do consumidor Nesta questão você não precisa resolver o modelo apenas interpretar as equações que serão apresentadas a 075 ponto Ao resolver o problema do consumidor uma das equações que se pode obter é c0y α 1α γ0y Considerando esta equação e a função de utilidade interprete o significado da constante γ0 relacionando com a Lei de Engel discutida na questão 1 b 1 ponto A solução completa do modelo implica que sob livre comércio a participação dos serviços na economia doméstica será igual a LS 1 α2 AM AM 1 1 βM βSσ AS AM1 σ onde AM é a produtividade do trabalho do setor de manufaturados na economia estrangeira Compare essa expressão com a obtida ao final da questão 2 e para o caso em que a economia estrangeira é mais produtiva no setor de manufaturas AM AM explique como a transformação estrutural do país doméstico é afetada pelo país estrangeiro c 075 ponto Rodrik 2016 e outros autores argumentam que a abertura comercial dos anos 1990 teria provocado efeitos distintos na América Latina e na Ásia Explique quais seriam esses efeitos envolvendo o conceito de desindustrialização prematura Transformação Estrutural Desenvolvimento SócioEconômico II Thiago Sevilhano Martinez FCE UERJ 3 de dezembro de 2025 Transformação Estrutural Kuznets 1966 Modern Economic Growth We identify the economic growth of nations as a sustained increase in per capita or per worker product most often accompanied by an increase in population and usually by sweeping structural changes In modern economic times these were changes in the industrial structure within which product was turned out and resources employed away from agriculture toward nonagricultural activities the process of industrialization in the distribution of population between the countryside and the cities the process of urbanization in the relative economic position of groups within the nation distinguished by employment status attachment to various industries level of per capita income and the like 1 56 Transformação Estrutural Clássicos do desenvolvimento econômico Hirschman Nurkse RosensteinRodan Lewis Estruturalismo latinoamericano CEPAL Prebisch Furtado e outros Literatura recente modelos de crescimento e desenvolvimento para explicar movimentos de longo prazo na composição setorial das economias Algumas resenhas desta literatura recente Matsuyama 2008 Herrendorf Rogerson e Valentinyi 2014 Alessandria Johnson e Yi 2023 Em livrotexto Acemoglu 2009 cap 20 2 56 Conceito Na literatura contemporânea mainstream de crescimento e desenvolvimento econômico Transformação estrutural processo histórico de mudança na importância relativa de setores econômicos Importância dos setores em termos de Uso de fatores de produção ex participação no emprego Padrões de consumo Composição do valor adicionado 3 56 Setores Corte setorial Mais frequente Agricultura Indústria e Serviços Outros possíveis mais desagregados ex serviços modernos x tradicionais saúde x não saúde Padrão típico economias maduras Agricultura queda de participação Indústria trajetória de U invertido Serviços aumento de participação 4 56 Padrões históricos EUA Acemoglu 2009 p698 5 56 Padrões históricos Agricultura Herrendorf Rogerson e Valentinyi 2014 6 56 Padrões históricos Indústria Herrendorf Rogerson e Valentinyi 2014 7 56 Padrões históricos Serviços Herrendorf Rogerson e Valentinyi 2014 8 56 Padrões históricos Vários países 9 56 Determinantes Principais razões da transformação estrutural 1 Lado da demanda efeito renda lei de Engel 2 Lado da oferta efeito preço relativo Baumol 1967 3 Inserção no comércio internacional Matsuyama 2009 10 56 Lei de Engel Mudança no padrão de consumo conforme a renda cresce Em particular queda da participação dos alimentos no gasto Consequência redução da participação da agricultura no PIB 11 56 Lei de Engel CMAP 2022 Desoneração de PISCOFINS sobre os produtos da Cesta Básica Dados da Pesquisa Nacional de Orçamentos Familiares POF 20022003 20082009 e 20172018 12 56 Curva de Engel Preferências homotéticas Preferências homotéticas A curva de rendaconsumo A e a curva de Engel B no caso de preferências homotéticas Varian cap 6 13 56 Curva de Engel Preferências nãohomotéticas Preferências StoneGeary uCA CM CS CA γAαACMαMCS γSαS com αA αM αS 1 Forma mais simples de modelar lei de Engel Consumo mínimo de sobrevivência de alimentos γA Dotaçãode serviços sem precisar comprar γS Implicação crescimento desbalanceado Kongsamut Rebelo e Xie 2001 14 56 Efeito Baumol Baumol 1967 Macroeconomics of Unbalanced Growth The Anatomy of Urban Crisis The American Economic Review Hipótese principal dois grupos de atividades i Setor progressivo inovações acumulação de capital economias de escala aumentos recorrentes da produtividade por hora trabalhada ex indústria de transformação ii Setor nãoprogressivo atividades que só permitem aumentos esporádicos da produtividade ex diversos serviços como educação arte e outros 15 56 Efeito Baumol Outras hipóteses assumptions simplicadoras i Único fator de produção é o trabalho ii Mesmo salário nos dois setores iii Salário acompanha crescimento da produtividade no setor dinâmico 16 56 Efeito Baumol Resultados Baumol 1967 1 Custo unitário constante no setor progressivo crescente no setor nãoprogressivo 2 Demanda no setor nãoprogessivo se elástica à renda ou inelástica ao preço não há redução no consumo 3 Alocação do trabalho mantendose proporção constante na quantidade produzida dos dois bens fração crescente da mãodeobra se desloca ao setor nãoprogressivo emprego no setor progressivo tende a zero 4 Crescimento econômico crescimento do produto por trabalhador tende a zero 17 56 Efeito Baumol Literatura recente 1 Efeito preçorelativo crescimento da produtividade maior na indústria é que nos serviços salário médio equiparado aumento de custo relativo nos serviços aumento de preço relativo dos serviços complementariedade na demanda por indústria e serviços aumenta participação dos serviços na economia cai participação da indústria 2 Ngai e Pissarides 2007 incorporam efeito Baumol em modelo de crescimento contemporânero 3 Herrendorf Rogerson e Valentinyi 2014 conforme vários artigos é a principal razão para a desindustrialização das economias maduras 18 56 Elasticidade de substituição Efeito Baumol Seja a função de utilidade uc1 c2 Qual é a variação proporcional da demanda relativa c1c2 em resposta a certo aumento proporcional no preço relativo p1p2 dos bens c1c2 c1c2 p1p2 p1p2 Elasticidade de substituição σ c1c2 p1p2 p1p2 c1c2 ou em logs σ logc1c2 logp1p2 19 56 Elasticidade de substituição Efeito Baumol Bens substitutos σ 1 variação proporcional na demanda relativa maior que variação proporcional nos preços relativos Δc1 c2 c1 c2 Δp1 p2 p1 p2 Bens complementares σ 1 variação proporcional na demanda relativa menor que variação proporcional nos preços relativos Δc1 c2 c1 c2 Δp1 p2 p1 p2 Elasticidade de substituição Efeito Baumol Caso CobbDouglas σ 1 variação proporcional na demanda relativa igual a variação proporcional nos preços relativos Δc1 c2 c1 c2 Δp1 p2 p1 p2 Seja um consumidor com utilidade uc1 c2 c1β c21β e restrição orçamentária p1 c1 p2 c2 y Escolha ótima c1 c2 β 1 β p2 p1 Logo σ logc1 c2 logp1 p2 1 pois logc1 c2 logβ 1 β logp1 p2 Elasticidade de substituição Efeito Baumol Caso CobbDouglas Interpretação sob escolha ótima com utilidade CobbDouglas proporção dos gastos em cada bem é sempre constante p1c1 p2c2 β 1 β Assim se dobrar o preço relativo p1 p2 a demanda relativa c1 c2 cai pela metade mantendo a proporção de gasto β 1β entre os bens 22 56 Elasticidade de substituição Efeito Baumol Utilidade CES Constant Elasticity of Substitution CES família de funções de utilidade em que a elasticidade de substituição entre os bens é constante Forma funcional uc1 c2 β1 c1θ β2 c2θ1θ A escolha ótima será c1c2 β1β2 p2p111θ e a elasticidade de substituição é constante igual a σ 11 θ Elasticidade de substituição Efeito Baumol Utilidade CES casos especiais CobbDouglas σ 1 ou θ 0 uc1 c2 c1β1c2β2 Complementares perfeitos σ 0 ou θ uc1 c2 minc1 c2 Substitutos perfeitos σ ou θ 1 uc1 c2 β1c1 β2c2 24 56 Efeito Baumol Modelo Adaptado de Matsuyama 2009 Versão do modelo ricardiano de Matsuyama 2009 com economia fechada e dois bens de consumo manufatura M e serviços S Um fator de produção oferta inelástica trabalho LM LS 1 Dotação unitária de trabalho LM e LS são participações no emprego 25 56 Efeito Baumol Modelo Famílias Utilidade CES ucM cS βM cMθ βS cSθ1θ 1 Problema das famílias escolher cM e cS para maximizar ucM cS sujeitas à restrição orçamentária pM cM pS cS y 2 onde a renda da família é y LM W LS W W Efeito Baumol Modelo Famílias maximização de utilidade Função Lagrangiana L ucM cS λpM cM pS cS y Condições de primeira ordem Lci 0 para i M S logo 1θ uβM cMθ βS cSθ θ βi ciθ 1 λ pi Tomando λ pM λ pS chegase a cMcS βMβS pSpM11 θ 3 Efeito Baumol Modelo Famílias elasticidade de substituição Recordando a elasticidade de substituição é σ 1 1 θ 4 pois aplicando log à equação 3 log cM cS 1 1 θ log βM βS 1 1 θ log pM pS de maneira que usando a denição σ logcMcS logpMpS 1 chegase à relação 4 28 56 Efeito Baumol Modelo Famílias De 3 e 4 cMcS βMβS pSpMσ 5 ou pM cM pS cS βMβSσ pMpS1σ 6 Combinando 6 à restrição orçamentária 2 pS cS y 1 1 βMβSσ pMpS1σ 7 Efeito Baumol Modelo Firmas Firmas concorrenciais nos setores i M S Funções de produção YM AMLM YS ASLS 8 Preços pM e pS salário W nos dois setores Firmas escolhem Lj para maximizar lucros πi piYi WLi Solução W piAi logo pM pS AS AM 9 30 56 Efeito Baumol Modelo Market clearing Mercado de trabalho LM LS 1 10 Mercado de bens industrializados CM YM 11 Mercado de serviços CS YS 12 31 56 Efeito Baumol Modelo Equilíbrio Como y W e W pS AS juntando as equações 8 9 7 10 e 11 temos LS 1 1 βMβSσ ASAM1σ 13 Efeito Baumol se σ 1 cM e cS complementares progresso técnico mais rápido na indústria em comparação aos serviços queda em ASAM implica aumento em LS No limite ASAM 0 logo LS 1 e LM 0 Toda a mão de obra tende a trabalhar nos serviços emprego na indústria tende a zero Mudança estrutural economias abertas Determinantes 33 56 Gráfico 1 Participação da indústria no PIB do Brasil 19702022 Fonte Ipeadata séries concatenadas do PIB a preços básicos das contas nacionais do IBGE Indústria excluindo a construção civil Participações calculadas sobre a soma do PIB a preços básicos de agricultura indústria e serviços Mudança estrutural economias abertas Matsuyama 2009 Modelos de comércio internacional pouca ênfase em entender como comércio afeta mudança estrutural foco em explicar composição e volume de comércio impactos sob estrutura econômica dada Modelos de macroeconomia e desenvolvimento mudança estrutural em modelos de economia fechada análises de que comparam países supondo que são observações independentes A partir de Matsuyama 2009 artigos combinando as duas vertentes 35 56 Mudança estrutural economias abertas Rodrik 2016 Desindustrialização prematura países de renda baixa e média têm se tornado economias de serviço em níveis de renda per capita muito menores que os observados nas economias maduras América Latina é a região mais atingida por esse fenômeno Por não terem vantagens comparativas sólidas na indústria como os asiáticos ao abrir suas economias importaram a desindustrialização das economias avançadas pelas tendências de preços relativos de bens manufaturados 36 56 Mudança estrutural economias abertas Alessandria Johnson e Yi 2021 Resenha de literatura efeitos do comércio sobre trajetórias de mudança estrutural Modelos com estudos de caso economias maduras ou milagres asiáticos abertura aprofunda industrialização em com vantagens comparativas na indústria Ex Uy Yi e Zhang 2013 Coreia do Sul abertura gera aumento de 10pp na participação da indústria no emprego Carência de estudos economias com vantagens comparativas em agricultura e mineração 37 56 Mudança estrutural economias abertas Spozi Yi e Zhang 2022 Desindustrialização queda na participação da indústria no PIB mediana dos países Polarização industrial aumento da dispersão entre países na participação da indústria no PIB Ásia x América Latina 38 56 Participação da Indústria no PIB mediana e dispersão 28 países 19712011 Fonte Extraído de Sposi Yi e Zhang 2022 A linha sólida traça o valor mediano das parcelas de valor agregado da manufatura entre os países ao longo do tempo eixo y enquanto as bandas superior e inferior correspondem aos percentis 99 e 1 respectivamente Amostra de 28 países de renda média e alta Polarização industrial Variância na participação da indústria no PIB 28 países 19712011 Fonte Extraído de Sposi Yi e Zhang 2022 Variância do logaritmo natural da participação da indústria no PIB ao longo dos anos Mudança estrutural economias abertas Spozi Yi e Zhang 2022 Modelo ricardiano decomposição por cenários contrafactuais 28 países 19712011 Motivo da desindustrialização 60 efeito preço relativo 40 iteração efeito preço relativo abertura comercial Motivo da polarização industrial 100 abertura comercial 41 56 Mudança estrutural economias abertas Modelo Matsuyama 2009 Modelo ricardiano de Matsuyama 2009 com dois países doméstico e estrangeiro Três bens de consumo agricultura O manufatura M e serviços S Numerário agricultura pO p O 1 Comercializáveis com o exterior bens manufaturados M e agrícolas O Não comercializáveis serviços S 42 56 Mudança estrutural economias abertas Modelo Matsuyama 2009 M e S produzidos em cada país com fator trabalho de oferta inelástica e dotação unitária LM LS 1 e L M L S 1 Bem agrícola produzido diretamente a partir de uma unidade de recurso natural Dotação de recursos naturais a cada período R e R OBS equações referentes à economia estrangeira serão omitidas quando análogas à economia doméstica 43 56 Mudança estrutural economias abertas Modelo Famílias Utilidade ucO cM cS cO γOα βM cMθ βs cSθ1 αθ 14 Alimento O bem essencial requisito mínimo de sobrevivência γO Problema das famílias escolher cO cM e cS para maximizar ucO cM cS sujeitos à restrição orçamentária cO pM cM pS cS y 15 onde a renda da família é y R LM W LS W R W Mudança estrutural economias abertas Modelo Famílias maximização de utilidade Função Lagrangiana L ucO cM cS λcO pMcM pScS y Condição de primeira ordem L cO 0 logo αu cO γO λ 16 Condições de primeira ordem L ci 0 para i M S logo 1 αu βMcMθ βScS θβici θ1 λpi 17 45 56 Mudança estrutural economias abertas Modelo Famílias Combinando as CPOs em 17 para M e S encontrase pS cS pM cM βS cSθ βM cMθ 18 ou cM cS βM βS pS pM σ e pM cM pS cs βM βS σ pM pS 1σ 19 onde σ 1 1θ é a elasticidade de substituição entre M e S Mudança estrutural economias abertas Modelo Famílias Combinando as CPOs para O e M 16 e 17 chegase a cO γO α1α 1 βS cSθ βM cMθ pM cM 20 que combinada a 18 implica cO γO α1α pM cM pS cS 21 Mudança estrutural economias abertas Modelo Famílias Substituindo na restrição orçamentária 15 a equação 21 para cO e depois a equação 19 para pM cM encontrase pS cS y 1 α 1 γO y 1 βM βS σ pM pS 1σ 22 Mudança estrutural economias abertas Modelo Firmas Firmas concorrenciais nos setores i M S Funções de produção YM AMLM YS ASLS 23 Preços pM e pS salário W nos dois setores Solução após maximização de lucros pi W Ai pM pS AS AM 24 49 56 Mudança estrutural economias abertas Modelo Market clearing economia aberta solução interior Mercado de trabalho LM LS 1 L M L S 1 25 Mercado de serviços CS YS C S Y S 26 Mercado de bens agrícolas CO C O R R 27 Mercado de bens industrializados CM C M YM Y M 28 e como PM P M de 24 W W A M AM 29 50 56 Mudança estrutural economias abertas Modelo Equilíbrio Matsuyama 2009 supõe R R Combinando as equações anteriores mostrase então que LS frac1fracalpha2 leftfracAMAM 1 right1leftfracbetaMbetaSrightsigma leftfracASAMright1sigma 30 LS frac1fracalpha2 leftfracAMAM 1 right1leftfracbetaMbetaSrightsigma leftfracASAMright1sigma 31 Mudança estrutural economias abertas Modelo Análise com σ1 Decomposição LS LS LS Primeiro termo igual a economia fechada efeito Baumol LS frac11 leftfracbetaMbetaSrightsigma leftfracASAMright1sigma 32 Efeito Baumol se sigma 1 cM e cS complementares progresso técnico mais rápido na indústria em comparação aos serviços queda em AS AM implica aumento em LS No limite AS AM o 0 logo LS o 1 e LM o 0 Toda a mão de obra tende a trabalhar nos serviços emprego na indústria tende a zero Mudança estrutural economias abertas Modelo Análise com σ1 Segundo termo LSII fracfracalpha2 left fracAMAM 1 right1 leftfracbetaMbetaSrightsigma leftfracASAMright1sigma 33 LSII fracfracalpha2 left fracAMAM 1 right1 leftfracbetaMbetaSrightsigma leftfracASAMright1sigma 34 Se AM AM país estrangeiro com vantagem comparativa na indústria exporta parte da desindustrialização que seria gerada pelo efeito Baumol em seu país Mudança estrutural economias abertas Modelo Heterogeneidade de dotação de recursos naturais E se R R Seja R vR com 0 v 1 Mostrase Martinez 2024 LS 1 α2 AMAM 1 1α2 1vW R 1 βMβSσ ASAM1σ 35 LS 1 α2 AMAM 1 1α2 1vW R AMAM 1 βMβSσ ASAM1σ 36 Se R R país doméstico abundante em recursos naturais importa parte da desindustrialização gerada pelo efeito Baumol no país estrangeiro Referências ALESSANDRIA George JOHNSON Robert YI KeiMu Perspectives on trade and structural transformation Cambridge NBER 2021 Working Paper n 28720 BAUMOL William Macroeconomics of unbalanced growth the anatomy of urban crisis The American Economic Review v 57 n 3 p 415426 1967 HERRENDORF Berthold ROGERSON Richard VALENTINYI Akos Growth and structural transformation In AGHION Philippe DURLAUF Steven Ed Handbook of economic growth Amsterdam Elsevier 2014 v 2 p 855941 MARTINEZ Thiago Sevilhano Mudança estrutural e abertura comercial no Brasil considerações a partir de modelos de equilíbrio geral In Anais do 52º Encontro Nacional de Economia da ANPEC Associação Nacional dos Centros de Pósgraduação em Economia 2024 55 56 Referências MATSUYAMA Kiminori Structural change in an interdependent world a global view of manufacturing decline Journal of the European Economic Association v 7 n 23 p 478486 2009 NGAI L Rachel PISSARIDES Christopher Structural change in a multisector model of growth American Economic Review v 97 n 1 p 429443 2007 RODRIK Dani Premature deindustrialization Journal of Economic Growth v 21 p 133 2016 SPOSI Michael YI KeiMu ZHANG Jing Deindustrialization and industry polarization Cambridge NBER 2022 Working Paper n 29483 Em revisão e resubmissão na Econometrica UY Timothy YI KeiMu ZHANG Jing Structural change in an open economy Journal of Monetary Economics v 60 n 6 p 667682 2013 56 56 QUESTÃO 1 A A Lei de Engel do consumo afirma que à medida que a renda de uma família aumenta a fração da renda ou do gasto total destinada a alimentos tende a diminuir mesmo que o gasto absoluto com alimentos possa aumentar Para formalizar essa ideia considere a renda ou despesa total de consumo Y e a despesa com alimentos EaY A participação dos alimentos no orçamento é sa Y Ea Y Y A Lei de Engel é a afirmação empírica de que essa participação é decrescente na renda d dY Ea Y Y 0 Isso é compatível com o fato de alimentos serem em geral bens de primeira necessidade a despesa com alimentos cresce com a renda mas menos do que proporcionalmente Em termos de elasticidaderenda da despesa com alimentos ηad Ea dY Y Ea e tipicamente0ηa1 Assim quando Y aumenta Ea pode aumentar porém saEa Y cai caracterizando exatamente o comportamento descrito pela Lei de Engel B No gráfico o eixo horizontal ordena as famílias por centésimos de renda familiar per capita do menor para o maior e o eixo vertical mostra a participação dos gastos com alimentos no total das despesas de consumo isto é sa Ea ET em que Ea é a despesa com alimentos e ET é a despesa total de consumo A Lei de Engel prevê que quando a renda ou o nível de despesa total aumenta a participação sa tende a cair Isso aparece no gráfico pelo declive negativo das curvas nos centésimos mais baixos de renda sa é alta na ordem de 30 a 40 e vai diminuindo conforme se caminha para os centésimos mais altos chegando perto de 10 a 15 no topo da distribuição Em termos de variação ao longo do eixo horizontal a leitura é Δ sa Δcentésimo de renda 0 ou seja famílias mais ricas comprometem uma parcela menor do orçamento com alimentos mesmo podendo gastar mais em valor absoluto Além disso para um mesmo centésimo de renda as curvas de anos mais recentes tendem a ficar abaixo das mais antigas por exemplo 2017 2018 abaixo de 20022003 em grande parte do intervalo Essa diferença vertical pode ser lida como uma redução da participação dos alimentos no orçamento ao longo do tempo consistente com melhora do poder de compra eou mudanças nos padrões de consumo mantendose a comparação por posição na distribuição a fração destinada a alimentos tornase menor enquanto outros itens passam a ocupar parcela relativamente maior do gasto total C A transformação estrutural é o processo pelo qual a economia conforme a renda per capita cresce realoca recursos trabalho capital e demanda entre setores tipicamente reduzindo o peso relativo do setor primário e em muitos casos também o da indústria em favor de serviços e de bens de maior valor agregado A Lei de Engel é um dos mecanismos de demanda que impulsionam esse processo Seja Y a renda e EaY a despesa com alimentos Com sa Y Ea Y Y e d sa dY 0 temse que à medida que Y aumenta a parcela do orçamento destinada a alimentos cai Como a soma das participações setoriais do gasto é 1 isto é sa Y so Y 1 em que so Y representa a participação conjunta de todos os demais bens e serviços segue que so Y 1sa Y d so dY d sa dY 0 Logo o crescimento da renda desloca a composição da demanda para fora de alimentos e para dentro de outros bens e principalmente serviços educação saúde transporte lazer habitação cuidados pessoais etc As firmas e os trabalhadores reagem a esse novo padrão de demanda setores ligados à alimentação agropecuária e parte da indústria de alimentos tendem a crescer menos em termos relativos enquanto setores que produzem os itens com participação crescente tendem a expandir sua produção investimento e emprego Esse mecanismo também pode ser expresso por elasticidadesrenda Se ηad Ea dY Y Ea 1eηod Eo dY Y Eo 1ou ao menos maior que ηa então com o aumento de Y o gasto em alimentos cresce menos do que proporcionalmente enquanto o gasto em outros bensserviços cresce mais ou relativamente mais aumentando seu peso no consumo A consequência macroeconômica é a realocação setorial a estrutura produtiva precisa se adaptar para atender a uma cesta de consumo mais diversificada e sofisticada o que caracteriza a transformação estrutural QUESTÃO 2 A A família representativa escolhe c M c S para maximizar sua utilidade CES u c McS β M cM θ βScS θ 1 θ sujeita à restrição orçamentária Como a dotação total de horas é unitária LM LS1 o salário por hora é o mesmo nos dois setores w e toda a renda vem do trabalho a renda total é ywLMLSw Como toda a renda é gasta em consumo de manufaturados e serviços a restrição orçamentária é pMc M pScSyw Como a função x x 1 θ é estritamente crescente no domínio relevante maximizar u c McS é equivalente a maximizar o seu agregador βM c M θ β ScS θ sob a mesma restrição O Lagrangiano pode ser escrito como L cM cS λβM cM θ βScS θ λ ypM cMpScS As condições de primeira ordem assumindo solução interior c M0 e cS0 são L cM βM θcM θ1λ pM0 L cS βSθcS θ1λ pS0 L λ ypM c MpScS0 Das duas primeiras equações λ βMθ cM θ1 pM e λ βSθcS θ1 pS Igualando as duas expressões para lambda βMθ cM θ1 pM βSθcS θ1 pS β M cM θ1 pM βScS θ1 pS Reorganizando βM pSc M θ1βS pM cS θ1 c M c S θ1 βS pM βM pS Como θ11θ cM cS 1θ βS pM β M pS cM cS 1θ βM pS βS pM Elevando ambos os lados à potência 1 1θ obtémse a relação pedida entre consumo relativo e preço relativo cM cS β M pS βS pM 1 1θ Essa expressão é a condição de ótimo que iguala a taxa marginal de substituição ao preço relativo implícita nas condições de primeira ordem acima B Partese da relação obtida no item a cM cS β M pS βS pM 1 1θ θ1 Tomando logaritmo natural em ambos os lados log cM cS 1 1θ log βM pS βS pM Separando os termos do logaritmo log cM cS 1 1θlog β M β S log pS pM Como log pS pMlog pM pS isso pode ser reescrito como log cM cS 1 1θ log βM βS 1 1θ log pM pS Agora derivando em relação a log pM pS log cM cS log pM pS 1 1θ pois log β M βS é constante não depende dos preços Pela definição dada no enunciado σ log cM cS log pM pS 1 1θ 1 1θ Como o resultado não depende de pM pS c M ou cS a elasticidade de substituição é de fato constante nesta economia o que caracteriza a forma CES C Do item a cM cS β M pS βS pM 1 1θ Como no item b foi obtido σ 1 1θ podese reescrever diretamente a demanda relativa como cM cS β M pS βS pM σ β M βS σ pS pM σ Logo c McS β M βS σ pS pM σ Substituise isso na restrição orçamentária pMc M pScSy pMcS βM βS σ pS pM σ pScS y Reorganizando o primeiro termo pMc S βM βS σ pS pM σ cS βM βS σ pS σ pM 1σ Então cS β M β S σ pS σ pM 1σ pScSy cS βM βS σ pS σ pM 1σ pSy Fatorando pS cS pS1 βM βS σ pS σ1 pM 1σy Como pS σ1 pM 1σ pM pS 1σ obtémse cS pS1 βM βS σ pM pS 1σ y e dividindo ambos os lados por y pScS y 1 1 βM βS σ pM pS 1σ Agora para analisar o efeito de um aumento de pS em relação a pM definese x pM pS Um aumento de pS pM implica queda de x A participação do gasto em serviços é pScS y 1 1 βM βS σ x 1σ Se σ1 substitutos então 1σ0 Com expoente negativo quando x cai x 1σ aumenta o denominador aumenta e portanto pScS y diminui A leitura econômica é que com alta substitutibilidade o encarecimento relativo de serviços leva a forte substituição para manufaturados reduzindo a participação do gasto com serviços na renda Se σ1 complementares então 1σ0 Com expoente positivo quando x cai x 1σ cai o denominador diminui e portanto pScS y aumenta A leitura econômica é que com baixa substitutibilidade complementaridade a família não consegue reduzir muito cS quando pS sobe relativamente assim a alta de preço pesa mais e a participação do gasto com serviços na renda tende a aumentar D Em concorrência perfeita cada firma toma preços e salário como dados No setor de manufaturados com tecnologia linear Y MAM LM o lucro é π MpM Y MW LMpM AM LM W LM pM AMW LM LM 0 Como a função objetivo é linear em LM a decisão ótima depende do sinal do coeficiente pM AMW π M LM pM AMW LM Se pM AMW 0 qualquer aumento de LM reduz o lucro então a firma escolhe LM 0 Se pM AMW 0 aumentar LM aumenta o lucro sem limite com essa tecnologia linear e sem restrição adicional o que é incompatível com um equilíbrio competitivo com escala finita Assim para que haja produção positiva e lucros não explodam deve valer a condição de lucro zero ou equivalentemente preço igual ao custo marginal pM AMW 0 pM W A M O mesmo raciocínio vale para o setor de serviços com Y SASLS e lucro πSpSY SW LSpS ASLSW LS pS ASW LSLS0 Para haver produção positiva em equilíbrio competitivo pS ASW 0 pS W AS Dividindo as duas expressões de preço pM pS W AM W AS AS AM Logo o preço relativo é determinado pela produtividade relativa pM pS AS A M E Do item c a participação do gasto com serviços na renda é pScS y 1 1 βM βS σ pM pS 1σ Do item d pM pS AS A M Substituindo pScS y 1 1 βM βS σ AS AM 1σ Agora entram as condições de market clearing e a tecnologia Em equilíbrio a produção de cada setor é consumida c MY McSY S Com as funções de produção Y MAM LM Y SAS LS logo cSASLS Além disso como a oferta total de trabalho é unitária LM LS1 a renda do consumidor é yW LMLSW Do item d o preço de serviços é pS W AS Então o lado esquerdo pode ser reescrito inteiramente em termos de LS pScS y W ASAS LS W LS Igualando com a expressão obtida acima chegase à fração da força de trabalho em serviços como função da produtividade relativa LS 1 1 βM βS σ AS AM 1σ Para apresentar o raciocínio de Baumol definese r AS AM e K βM βS σ 0 de modo que LS r 1 1K r 1σ Se a produtividade em manufaturas cresce mais rapidamente do que em serviços então AM sobe relativamente a AS o que implica queda de r AS AM ao longo do tempo Ao mesmo tempo pelo item d pS pM A M AS isto é serviços ficam relativamente mais caros quando AM avança mais do que AS que é a doença dos custos de Baumol O efeito dessa mudança de preços sobre a alocação de trabalho depende de σ Derivando LS r em relação a r d LS dr K 1σ r σ 1K r 1σ 2 Como K0 r σ0 e o denominador é positivo o sinal é determinado por 1σ Quando σ1 complementares temse 1σ0 então d LS dr 0 Assim quando r AS AM cai manufaturas ficam relativamente muito mais produtivas LS aumenta A interpretação é que a economia não consegue substituir consumo de serviços por manufaturados com facilidade como serviços ficam relativamente mais caros e sua produtividade cresce pouco é necessário deslocar mais trabalho para serviços para atender a demanda de equilíbrio reduzindo a participação relativa da indústria no emprego Quando σ1 substitutos temse 1σ0 então d LS dr 0 Assim quando r cai LS cai A interpretação é que com alta substitutibilidade o encarecimento relativo de serviços induz forte substituição em direção a manufaturados o que reduz a participação de serviços no gasto e portanto na alocação de trabalho enfraquecendo ou revertendo a tendência de terciarização associada ao efeito Baumol No caso limite σ1 CobbDouglas temse 1σ0 então r 1σ1 e LS 1 1 βM βS βS βMβS isto é a participação de trabalho em serviços fica constante independentemente de AS AM QUESTÃO 3 A A utilidade do consumidor doméstico é u c0cM cSc0γ0 αβ Mc M θ βScS θ 1α θ 0α1γ00 com restrição orçamentária c0 pM cMpScSy p01 O termo c0γ 0 implica uma preferência do tipo StoneGeary para o bem 0 para que a utilidade esteja bem definida no domínio usual consumo não negativo do excedente é necessário c0γ0 Assim γ 0 funciona como um consumo mínimo ou nível de subsistência do bem tradable extraído de recursos naturais somente o que excede γ 0 gera utilidade na parcela c0 A equação fornecida como condição da solução do consumidor é c0 y α1α γ 0 y Ela mostra que a participação do gasto no bem 0 na renda é composta de duas partes um componente assintótico α e um componente de subsistência proporcional a γ 0 y Como γ 0 y y 0 segue que c0 y y α Logo quanto maior a renda y menor é a importância relativa do termo de subsistência 1α γ0 y e a parcela do orçamento destinada a c0 converge para α Diferenciando a expressão dada em relação a y d dy c0 y 1α d dy γ 0 y 1α γ 0 y 2 1α γ 0 y 2 0 o que confirma que c0 y cai quando a renda aumenta Essa queda da participação orçamentária com a renda é exatamente o conteúdo da Lei de Engel para um bem de necessidade Aqui γ 0 é o parâmetro que introduz nãohomoteticidade e gera o padrão de Engel em níveis baixos de renda o componente de subsistência é relevante e a participação c0 y é maior à medida que y cresce esse componente perde peso e sobra relativamente mais renda para ser destinada aos demais bens c M c S B Ao final da questão 2 a fração de trabalho em serviços na economia fechada ficou determinada por LS fechada 1 1 βM βS σ AS AM 1σ No modelo com abertura comercial a expressão fornecida para a economia doméstica sob livre comércio é LS aberta 1 α 2 A M A M 1 1 β M β S σ AS AM 1σ Comparando as duas notase que o denominador é o mesmo do caso fechado e que a abertura introduz um fator adicional no numerador Isso pode ser escrito como LS aberta1 α 2 A M A M 1LS fechada No caso pedido em que a economia estrangeira é mais produtiva em manufaturas AM AM temse AM AM 10 1 α 2 AM AM 11 e portanto LS abertaLS fechada A interpretação econômica é que a maior produtividade estrangeira em manufaturas torna o setor manufatureiro relativamente mais baratoeficiente no exterior Com comércio a economia doméstica pode atender parte maior de sua demanda por manufaturados via importações ou equivalentemente enfrenta uma concorrência externa mais forte nesse setor o que reduz o incentivo de manter trabalho doméstico alocado em manufaturas Como serviços são nãotradables no modelo a realocação do trabalho ocorre principalmente na direção de serviços elevando LS e reduzindo a participação relativa da manufatura no emprego doméstico O parâmetro α regula a intensidade desse canal quanto maior alpha maior é o multiplicador 1 α 2 A M A M 1 e mais forte é o aumento de LS gerado pela vantagem de produtividade estrangeira em manufaturas Já σ continua sendo central porque determina o nível de base de LS via o denominador isto é como a demanda doméstica reage ao preço relativo ligado a AS AM Assim a abertura quando AM AM desloca a economia doméstica na direção de uma transformação estrutural com maior participação de serviços e o quanto isso aparece no equilíbrio depende também de quão facilmente manufaturados e serviços se substituem valor de σ C A desindustrialização pode ser entendida como a queda da participação da manufatura no produto e sobretudo no emprego isto é uma queda de sM em que sM representa a parcela do emprego ou do valor adicionado na manufatura Em muitas trajetórias históricas sM primeiro sobe com o aumento de renda per capita e depois cai quando a economia fica rica e os serviços ganham espaço isso pode ser representado de forma simplificada por uma relação em formato de U invertido entre sM e a renda y A desindustrialização prematura ocorre quando esse pico de sM acontece em um nível de renda y relativamente baixo e a queda começa antes de o país consolidar capacidades industriais e produtividade elevadas Em termos de trajetória significa que o máximo de sM y é atingido cedo e a economia passa a perder indústria quando ainda não alcançou o patamar de renda típico das economias que se industrializaram plenamente Rodrik 2016 e outros autores argumentam que a abertura comercial dos anos 1990 teria interagido com tendências globais e características regionais para gerar efeitos distintos na América Latina e na Ásia Um canal importante é a competição externa em manufaturas e a especialização induzida por vantagens comparativas e por preços relativos internacionais Se o país passa a importar manufaturados a preços relativamente menores eou enfrenta concorrência muito forte de grandes exportadores a demanda por manufaturas domésticas e a lucratividade do setor podem cair deslocando trabalho para serviços eou setores primários reduzindo sM Na Ásia a abertura foi frequentemente associada a uma estratégia de crescimento liderado por exportações de manufaturados e integração em cadeias globais de valor Com isso a demanda externa por manufaturas domésticas cresceu e a manufatura pôde absorver trabalho aprender elevar produtividade e diversificar a estrutura produtiva Nesse caso a abertura tende a deslocar a economia na direção de maior produção industrial ao menos por um período relevante sustentando sM por mais tempo e postergando a queda Na América Latina a abertura dos anos 1990 combinada em vários casos com apreciação cambial e posteriormente com ciclos de commodities frequentemente reforçou a especialização em bens primários e ampliou a penetração de importações manufatureiras Assim mesmo com ganhos de eficiência parte da indústria local perdeu participação relativa antes de atingir grande escala e sofisticação tecnológica enquanto serviços inclusive serviços de baixa produtividade absorveram parcela crescente do emprego Esse conjunto de forças é compatível com uma queda de sM em níveis de renda relativamente mais baixos caracterizando a desindustrialização prematura Além disso mesmo quando a produção manufatureira não cai tanto o progresso técnico tende a reduzir a intensidade de trabalho na manufatura de modo que o emprego industrial pode cair mais cedo do que o produto industrial Isso acelera a queda de sM do emprego reforçando o diagnóstico de prematuridade em economias que abrem cedo e enfrentam competição global intensa sem uma trajetória longa de expansão industrial QUESTÃO 1 A A Lei de Engel do consumo afirma que à medida que a renda de uma família aumenta a fração da renda ou do gasto total destinada a alimentos tende a diminuir mesmo que o gasto absoluto com alimentos possa aumentar Para formalizar essa ideia considere a renda ou despesa total de consumo Y e a despesa com alimentos 𝐸𝑎𝑌 A participação dos alimentos no orçamento é 𝑠𝑎𝑌 𝐸𝑎𝑌 𝑌 A Lei de Engel é a afirmação empírica de que essa participação é decrescente na renda 𝑑 𝑑𝑌 𝐸𝑎𝑌 𝑌 0 Isso é compatível com o fato de alimentos serem em geral bens de primeira necessidade a despesa com alimentos cresce com a renda mas menos do que proporcionalmente Em termos de elasticidaderenda da despesa com alimentos 𝜂𝑎 𝑑𝐸𝑎 𝑑𝑌 𝑌 𝐸𝑎 e tipicamente 0 𝜂𝑎 1 Assim quando Y aumenta 𝐸𝑎 pode aumentar porém 𝑠𝑎 𝐸𝑎 𝑌 cai caracterizando exatamente o comportamento descrito pela Lei de Engel B No gráfico o eixo horizontal ordena as famílias por centésimos de renda familiar per capita do menor para o maior e o eixo vertical mostra a participação dos gastos com alimentos no total das despesas de consumo isto é 𝑠𝑎 𝐸𝑎 𝐸𝑇 em que 𝐸𝑎 é a despesa com alimentos e 𝐸𝑇 é a despesa total de consumo A Lei de Engel prevê que quando a renda ou o nível de despesa total aumenta a participação 𝑠𝑎 tende a cair Isso aparece no gráfico pelo declive negativo das curvas nos centésimos mais baixos de renda 𝑠𝑎 é alta na ordem de 30 a 40 e vai diminuindo conforme se caminha para os centésimos mais altos chegando perto de 10 a 15 no topo da distribuição Em termos de variação ao longo do eixo horizontal a leitura é Δ𝑠𝑎 Δcentésimo de renda 0 ou seja famílias mais ricas comprometem uma parcela menor do orçamento com alimentos mesmo podendo gastar mais em valor absoluto Além disso para um mesmo centésimo de renda as curvas de anos mais recentes tendem a ficar abaixo das mais antigas por exemplo 20172018 abaixo de 20022003 em grande parte do intervalo Essa diferença vertical pode ser lida como uma redução da participação dos alimentos no orçamento ao longo do tempo consistente com melhora do poder de compra eou mudanças nos padrões de consumo mantendose a comparação por posição na distribuição a fração destinada a alimentos tornase menor enquanto outros itens passam a ocupar parcela relativamente maior do gasto total C A transformação estrutural é o processo pelo qual a economia conforme a renda per capita cresce realoca recursos trabalho capital e demanda entre setores tipicamente reduzindo o peso relativo do setor primário e em muitos casos também o da indústria em favor de serviços e de bens de maior valor agregado A Lei de Engel é um dos mecanismos de demanda que impulsionam esse processo Seja Y a renda e 𝐸𝑎𝑌 a despesa com alimentos Com 𝑠𝑎𝑌 𝐸𝑎𝑌 𝑌 e 𝑑𝑠𝑎 𝑑𝑌 0 temse que à medida que Y aumenta a parcela do orçamento destinada a alimentos cai Como a soma das participações setoriais do gasto é 1 isto é 𝑠𝑎𝑌 𝑠𝑜𝑌 1 em que 𝑠𝑜𝑌 representa a participação conjunta de todos os demais bens e serviços segue que 𝑠𝑜𝑌 1 𝑠𝑎𝑌 𝑑𝑠𝑜 𝑑𝑌 𝑑𝑠𝑎 𝑑𝑌 0 Logo o crescimento da renda desloca a composição da demanda para fora de alimentos e para dentro de outros bens e principalmente serviços educação saúde transporte lazer habitação cuidados pessoais etc As firmas e os trabalhadores reagem a esse novo padrão de demanda setores ligados à alimentação agropecuária e parte da indústria de alimentos tendem a crescer menos em termos relativos enquanto setores que produzem os itens com participação crescente tendem a expandir sua produção investimento e emprego Esse mecanismo também pode ser expresso por elasticidadesrenda Se 𝜂𝑎 𝑑𝐸𝑎 𝑑𝑌 𝑌 𝐸𝑎 1 e 𝜂𝑜 𝑑𝐸𝑜 𝑑𝑌 𝑌 𝐸𝑜 1 ou ao menos maior que 𝜂𝑎 então com o aumento de Y o gasto em alimentos cresce menos do que proporcionalmente enquanto o gasto em outros bensserviços cresce mais ou relativamente mais aumentando seu peso no consumo A consequência macroeconômica é a realocação setorial a estrutura produtiva precisa se adaptar para atender a uma cesta de consumo mais diversificada e sofisticada o que caracteriza a transformação estrutural QUESTÃO 2 A A família representativa escolhe 𝑐𝑀 𝑐𝑆 para maximizar sua utilidade CES 𝑢𝑐𝑀 𝑐𝑆 𝛽𝑀𝑐𝑀 𝜃 𝛽𝑆𝑐𝑆 𝜃 1 𝜃 sujeita à restrição orçamentária Como a dotação total de horas é unitária 𝐿𝑀 𝐿𝑆 1 o salário por hora é o mesmo nos dois setores w e toda a renda vem do trabalho a renda total é 𝑦 𝑤 𝐿𝑀 𝐿𝑆 𝑤 Como toda a renda é gasta em consumo de manufaturados e serviços a restrição orçamentária é 𝑝𝑀𝑐𝑀 𝑝𝑆𝑐𝑆 𝑦 𝑤 Como a função 𝑥 𝑥 1 𝜃 é estritamente crescente no domínio relevante maximizar 𝑢𝑐𝑀 𝑐𝑆 é equivalente a maximizar o seu agregador 𝛽𝑀𝑐𝑀 𝜃 𝛽𝑆𝑐𝑆 𝜃 sob a mesma restrição O Lagrangiano pode ser escrito como ℒ𝑐𝑀 𝑐𝑆 𝜆 𝛽𝑀𝑐𝑀 𝜃 𝛽𝑆𝑐𝑆 𝜃 𝜆𝑦 𝑝𝑀𝑐𝑀 𝑝𝑆𝑐𝑆 As condições de primeira ordem assumindo solução interior 𝑐𝑀 0 e 𝑐𝑆 0 são ℒ 𝑐𝑀 𝛽𝑀𝜃𝑐𝑀 𝜃1 𝜆𝑝𝑀 0 ℒ 𝑐𝑆 𝛽𝑆𝜃𝑐𝑆 𝜃1 𝜆𝑝𝑆 0 ℒ 𝜆 𝑦 𝑝𝑀𝑐𝑀 𝑝𝑆𝑐𝑆 0 Das duas primeiras equações 𝜆 𝛽𝑀𝜃𝑐𝑀 𝜃1 𝑝𝑀 e 𝜆 𝛽𝑆𝜃𝑐𝑆 𝜃1 𝑝𝑆 Igualando as duas expressões para lambda 𝛽𝑀𝜃𝑐𝑀 𝜃1 𝑝𝑀 𝛽𝑆𝜃𝑐𝑆 𝜃1 𝑝𝑆 𝛽𝑀𝑐𝑀 𝜃1 𝑝𝑀 𝛽𝑆𝑐𝑆 𝜃1 𝑝𝑆 Reorganizando 𝛽𝑀𝑝𝑆𝑐𝑀 𝜃1 𝛽𝑆𝑝𝑀𝑐𝑆 𝜃1 𝑐𝑀 𝑐𝑆 𝜃1 𝛽𝑆𝑝𝑀 𝛽𝑀𝑝𝑆 Como 𝜃 1 1 𝜃 𝑐𝑀 𝑐𝑆 1𝜃 𝛽𝑆𝑝𝑀 𝛽𝑀𝑝𝑆 𝑐𝑀 𝑐𝑆 1𝜃 𝛽𝑀𝑝𝑆 𝛽𝑆𝑝𝑀 Elevando ambos os lados à potência 1 1𝜃 obtémse a relação pedida entre consumo relativo e preço relativo 𝑐𝑀 𝑐𝑆 𝛽𝑀𝑝𝑆 𝛽𝑆𝑝𝑀 1 1𝜃 Essa expressão é a condição de ótimo que iguala a taxa marginal de substituição ao preço relativo implícita nas condições de primeira ordem acima B Partese da relação obtida no item a 𝑐𝑀 𝑐𝑆 𝛽𝑀𝑝𝑆 𝛽𝑆𝑝𝑀 1 1𝜃 𝜃 1 Tomando logaritmo natural em ambos os lados log 𝑐𝑀 𝑐𝑆 1 1 𝜃 log 𝛽𝑀𝑝𝑆 𝛽𝑆𝑝𝑀 Separando os termos do logaritmo log 𝑐𝑀 𝑐𝑆 1 1 𝜃 log 𝛽𝑀 𝛽𝑆 log 𝑝𝑆 𝑝𝑀 Como log 𝑝𝑆 𝑝𝑀 log 𝑝𝑀 𝑝𝑆 isso pode ser reescrito como log 𝑐𝑀 𝑐𝑆 1 1 𝜃 log 𝛽𝑀 𝛽𝑆 1 1 𝜃 log 𝑝𝑀 𝑝𝑆 Agora derivando em relação a log 𝑝𝑀 𝑝𝑆 log 𝑐𝑀 𝑐𝑆 log 𝑝𝑀 𝑝𝑆 1 1 𝜃 pois log 𝛽𝑀 𝛽𝑆 é constante não depende dos preços Pela definição dada no enunciado 𝜎 log 𝑐𝑀 𝑐𝑆 log 𝑝𝑀 𝑝𝑆 1 1 𝜃 1 1 𝜃 Como o resultado não depende de 𝑝𝑀 𝑝𝑆 𝑐𝑀 ou 𝑐𝑆 a elasticidade de substituição é de fato constante nesta economia o que caracteriza a forma CES C Do item a 𝑐𝑀 𝑐𝑆 𝛽𝑀𝑝𝑆 𝛽𝑆𝑝𝑀 1 1𝜃 Como no item b foi obtido 𝜎 1 1 𝜃 podese reescrever diretamente a demanda relativa como 𝑐𝑀 𝑐𝑆 𝛽𝑀𝑝𝑆 𝛽𝑆𝑝𝑀 𝜎 𝛽𝑀 𝛽𝑆 𝜎 𝑝𝑆 𝑝𝑀 𝜎 Logo 𝑐𝑀 𝑐𝑆 𝛽𝑀 𝛽𝑆 𝜎 𝑝𝑆 𝑝𝑀 𝜎 Substituise isso na restrição orçamentária 𝑝𝑀𝑐𝑀 𝑝𝑆𝑐𝑆 𝑦 𝑝𝑀 𝑐𝑆 𝛽𝑀 𝛽𝑆 𝜎 𝑝𝑆 𝑝𝑀 𝜎 𝑝𝑆𝑐𝑆 𝑦 Reorganizando o primeiro termo 𝑝𝑀𝑐𝑆 𝛽𝑀 𝛽𝑆 𝜎 𝑝𝑆 𝑝𝑀 𝜎 𝑐𝑆 𝛽𝑀 𝛽𝑆 𝜎 𝑝𝑆 𝜎𝑝𝑀 1𝜎 Então 𝑐𝑆 𝛽𝑀 𝛽𝑆 𝜎 𝑝𝑆 𝜎𝑝𝑀 1𝜎 𝑝𝑆𝑐𝑆 𝑦 𝑐𝑆 𝛽𝑀 𝛽𝑆 𝜎 𝑝𝑆 𝜎𝑝𝑀 1𝜎 𝑝𝑆 𝑦 Fatorando 𝑝𝑆 𝑐𝑆𝑝𝑆 1 𝛽𝑀 𝛽𝑆 𝜎 𝑝𝑆 𝜎1𝑝𝑀 1𝜎 𝑦 Como 𝑝𝑆 𝜎1𝑝𝑀 1𝜎 𝑝𝑀 𝑝𝑆 1𝜎 obtémse 𝑐𝑆𝑝𝑆 1 𝛽𝑀 𝛽𝑆 𝜎 𝑝𝑀 𝑝𝑆 1𝜎 𝑦 e dividindo ambos os lados por y 𝑝𝑆𝑐𝑆 𝑦 1 1 𝛽𝑀 𝛽𝑆 𝜎 𝑝𝑀 𝑝𝑆 1𝜎 Agora para analisar o efeito de um aumento de 𝑝𝑆 em relação a 𝑝𝑀 definese 𝑥 𝑝𝑀 𝑝𝑆 Um aumento de 𝑝𝑆 𝑝𝑀 implica queda de x A participação do gasto em serviços é 𝑝𝑆𝑐𝑆 𝑦 1 1 𝛽𝑀 𝛽𝑆 𝜎 𝑥1𝜎 Se 𝜎 1 substitutos então 1 𝜎 0 Com expoente negativo quando x cai 𝑥1𝜎 aumenta o denominador aumenta e portanto 𝑝𝑆𝑐𝑆 𝑦 diminui A leitura econômica é que com alta substitutibilidade o encarecimento relativo de serviços leva a forte substituição para manufaturados reduzindo a participação do gasto com serviços na renda Se 𝜎 1 complementares então 1 𝜎 0 Com expoente positivo quando x cai 𝑥1𝜎 cai o denominador diminui e portanto 𝑝𝑆𝑐𝑆 𝑦 aumenta A leitura econômica é que com baixa substitutibilidade complementaridade a família não consegue reduzir muito 𝑐𝑆 quando 𝑝𝑆 sobe relativamente assim a alta de preço pesa mais e a participação do gasto com serviços na renda tende a aumentar D Em concorrência perfeita cada firma toma preços e salário como dados No setor de manufaturados com tecnologia linear 𝑌𝑀 𝐴𝑀𝐿𝑀 o lucro é 𝜋𝑀 𝑝𝑀𝑌𝑀 𝑊𝐿𝑀 𝑝𝑀 𝐴𝑀𝐿𝑀 𝑊𝐿𝑀 𝑝𝑀𝐴𝑀 𝑊 𝐿𝑀 𝐿𝑀 0 Como a função objetivo é linear em 𝐿𝑀 a decisão ótima depende do sinal do coeficiente 𝑝𝑀𝐴𝑀 𝑊 𝜋𝑀𝐿𝑀 𝑝𝑀𝐴𝑀 𝑊𝐿𝑀 Se 𝑝𝑀𝐴𝑀 𝑊 0 qualquer aumento de 𝐿𝑀 reduz o lucro então a firma escolhe 𝐿𝑀 0 Se 𝑝𝑀𝐴𝑀 𝑊 0 aumentar 𝐿𝑀 aumenta o lucro sem limite com essa tecnologia linear e sem restrição adicional o que é incompatível com um equilíbrio competitivo com escala finita Assim para que haja produção positiva e lucros não explodam deve valer a condição de lucro zero ou equivalentemente preço igual ao custo marginal 𝑝𝑀𝐴𝑀 𝑊 0 𝑝𝑀 𝑊 𝐴𝑀 O mesmo raciocínio vale para o setor de serviços com 𝑌𝑆 𝐴𝑆𝐿𝑆 e lucro 𝜋𝑆 𝑝𝑆𝑌𝑆 𝑊𝐿𝑆 𝑝𝑆 𝐴𝑆𝐿𝑆 𝑊𝐿𝑆 𝑝𝑆𝐴𝑆 𝑊 𝐿𝑆 𝐿𝑆 0 Para haver produção positiva em equilíbrio competitivo 𝑝𝑆𝐴𝑆 𝑊 0 𝑝𝑆 𝑊 𝐴𝑆 Dividindo as duas expressões de preço 𝑝𝑀 𝑝𝑆 𝑊 𝐴𝑀 𝑊 𝐴𝑆 𝐴𝑆 𝐴𝑀 Logo o preço relativo é determinado pela produtividade relativa 𝑝𝑀 𝑝𝑆 𝐴𝑆 𝐴𝑀 E Do item c a participação do gasto com serviços na renda é 𝑝𝑆𝑐𝑆 𝑦 1 1 𝛽𝑀 𝛽𝑆 𝜎 𝑝𝑀 𝑝𝑆 1𝜎 Do item d 𝑝𝑀 𝑝𝑆 𝐴𝑆 𝐴𝑀 Substituindo 𝑝𝑆𝑐𝑆 𝑦 1 1 𝛽𝑀 𝛽𝑆 𝜎 𝐴𝑆 𝐴𝑀 1𝜎 Agora entram as condições de market clearing e a tecnologia Em equilíbrio a produção de cada setor é consumida 𝑐𝑀 𝑌𝑀 𝑐𝑆 𝑌𝑆 Com as funções de produção 𝑌𝑀 𝐴𝑀𝐿𝑀 𝑌𝑆 𝐴𝑆𝐿𝑆 logo 𝑐𝑆 𝐴𝑆𝐿𝑆 Além disso como a oferta total de trabalho é unitária 𝐿𝑀 𝐿𝑆 1 a renda do consumidor é 𝑦 𝑊 𝐿𝑀 𝐿𝑆 𝑊 Do item d o preço de serviços é 𝑝𝑆 𝑊 𝐴𝑆 Então o lado esquerdo pode ser reescrito inteiramente em termos de 𝐿𝑆 𝑝𝑆𝑐𝑆 𝑦 𝑊 𝐴𝑆 𝐴𝑆𝐿𝑆 𝑊 𝐿𝑆 Igualando com a expressão obtida acima chegase à fração da força de trabalho em serviços como função da produtividade relativa 𝐿𝑆 1 1 𝛽𝑀 𝛽𝑆 𝜎 𝐴𝑆 𝐴𝑀 1𝜎 Para apresentar o raciocínio de Baumol definese 𝑟 𝐴𝑆 𝐴𝑀 e 𝐾 𝛽𝑀 𝛽𝑆 𝜎 0 de modo que 𝐿𝑆𝑟 1 1 𝐾𝑟1𝜎 Se a produtividade em manufaturas cresce mais rapidamente do que em serviços então 𝐴𝑀 sobe relativamente a 𝐴𝑆 o que implica queda de 𝑟 𝐴𝑆 𝐴𝑀 ao longo do tempo Ao mesmo tempo pelo item d 𝑝𝑆 𝑝𝑀 𝐴𝑀 𝐴𝑆 isto é serviços ficam relativamente mais caros quando 𝐴𝑀 avança mais do que 𝐴𝑆 que é a doença dos custos de Baumol O efeito dessa mudança de preços sobre a alocação de trabalho depende de σ Derivando 𝐿𝑆𝑟 em relação a r 𝑑𝐿𝑆 𝑑𝑟 𝐾1 𝜎 𝑟𝜎 1 𝐾𝑟1𝜎2 Como 𝐾 0 𝑟𝜎 0 e o denominador é positivo o sinal é determinado por 1 𝜎 Quando 𝜎 1 complementares temse 1 𝜎 0 então 𝑑𝐿𝑆 𝑑𝑟 0 Assim quando 𝑟 𝐴𝑆 𝐴𝑀 cai manufaturas ficam relativamente muito mais produtivas 𝐿𝑆 aumenta A interpretação é que a economia não consegue substituir consumo de serviços por manufaturados com facilidade como serviços ficam relativamente mais caros e sua produtividade cresce pouco é necessário deslocar mais trabalho para serviços para atender a demanda de equilíbrio reduzindo a participação relativa da indústria no emprego Quando 𝜎 1 substitutos temse 1 𝜎 0 então 𝑑𝐿𝑆 𝑑𝑟 0 Assim quando r cai 𝐿𝑆 cai A interpretação é que com alta substitutibilidade o encarecimento relativo de serviços induz forte substituição em direção a manufaturados o que reduz a participação de serviços no gasto e portanto na alocação de trabalho enfraquecendo ou revertendo a tendência de terciarização associada ao efeito Baumol No caso limite 𝜎 1 CobbDouglas temse 1 𝜎 0 então 𝑟1𝜎 1 e 𝐿𝑆 1 1 𝛽𝑀 𝛽𝑆 𝛽𝑆 𝛽𝑀 𝛽𝑆 isto é a participação de trabalho em serviços fica constante independentemente de 𝐴𝑆 𝐴𝑀 QUESTÃO 3 A A utilidade do consumidor doméstico é 𝑢𝑐0 𝑐𝑀 𝑐𝑆 𝑐0 𝛾0𝛼𝛽𝑀𝑐𝑀 𝜃 𝛽𝑆𝑐𝑆 𝜃 1𝛼 𝜃 0 𝛼 1 𝛾0 0 com restrição orçamentária 𝑐0 𝑝𝑀𝑐𝑀 𝑝𝑆𝑐𝑆 𝑦 𝑝0 1 O termo 𝑐0 𝛾0 implica uma preferência do tipo StoneGeary para o bem 0 para que a utilidade esteja bem definida no domínio usual consumo não negativo do excedente é necessário 𝑐0 𝛾0 Assim 𝛾0 funciona como um consumo mínimo ou nível de subsistência do bem tradable extraído de recursos naturais somente o que excede 𝛾0 gera utilidade na parcela 𝑐0 A equação fornecida como condição da solução do consumidor é 𝑐0 𝑦 𝛼 1 𝛼 𝛾0 𝑦 Ela mostra que a participação do gasto no bem 0 na renda é composta de duas partes um componente assintótico α e um componente de subsistência proporcional a 𝛾0 𝑦 Como 𝛾0 𝑦 𝑦 0 segue que 𝑐0 𝑦 𝑦 𝛼 Logo quanto maior a renda y menor é a importância relativa do termo de subsistência 1 𝛼 𝛾0 𝑦 e a parcela do orçamento destinada a 𝑐0 converge para α Diferenciando a expressão dada em relação a y 𝑑 𝑑𝑦 𝑐0 𝑦 1 𝛼 𝑑 𝑑𝑦 𝛾0 𝑦 1 𝛼 𝛾0 𝑦2 1 𝛼𝛾0 𝑦2 0 o que confirma que 𝑐0 𝑦 cai quando a renda aumenta Essa queda da participação orçamentária com a renda é exatamente o conteúdo da Lei de Engel para um bem de necessidade Aqui 𝛾0 é o parâmetro que introduz nãohomoteticidade e gera o padrão de Engel em níveis baixos de renda o componente de subsistência é relevante e a participação 𝑐0 𝑦 é maior à medida que y cresce esse componente perde peso e sobra relativamente mais renda para ser destinada aos demais bens 𝑐𝑀 𝑐𝑆 B Ao final da questão 2 a fração de trabalho em serviços na economia fechada ficou determinada por 𝐿𝑆 fechada 1 1 𝛽𝑀 𝛽𝑆 𝜎 𝐴𝑆 𝐴𝑀 1𝜎 No modelo com abertura comercial a expressão fornecida para a economia doméstica sob livre comércio é 𝐿𝑆 aberta 1 𝛼 2 𝐴𝑀 𝐴𝑀 1 1 𝛽𝑀 𝛽𝑆 𝜎 𝐴𝑆 𝐴𝑀 1𝜎 Comparando as duas notase que o denominador é o mesmo do caso fechado e que a abertura introduz um fator adicional no numerador Isso pode ser escrito como 𝐿𝑆 aberta 1 𝛼 2 𝐴𝑀 𝐴𝑀 1 𝐿𝑆 fechada No caso pedido em que a economia estrangeira é mais produtiva em manufaturas 𝐴𝑀 𝐴𝑀 temse 𝐴𝑀 𝐴𝑀 1 0 1 𝛼 2 𝐴𝑀 𝐴𝑀 1 1 e portanto 𝐿𝑆 aberta 𝐿𝑆 fechada A interpretação econômica é que a maior produtividade estrangeira em manufaturas torna o setor manufatureiro relativamente mais baratoeficiente no exterior Com comércio a economia doméstica pode atender parte maior de sua demanda por manufaturados via importações ou equivalentemente enfrenta uma concorrência externa mais forte nesse setor o que reduz o incentivo de manter trabalho doméstico alocado em manufaturas Como serviços são não tradables no modelo a realocação do trabalho ocorre principalmente na direção de serviços elevando 𝐿𝑆 e reduzindo a participação relativa da manufatura no emprego doméstico O parâmetro 𝛼 regula a intensidade desse canal quanto maior alpha maior é o multiplicador 1 𝛼 2 𝐴𝑀 𝐴𝑀 1 e mais forte é o aumento de 𝐿𝑆 gerado pela vantagem de produtividade estrangeira em manufaturas Já σ continua sendo central porque determina o nível de base de 𝐿𝑆 via o denominador isto é como a demanda doméstica reage ao preço relativo ligado a 𝐴𝑆 𝐴𝑀 Assim a abertura quando 𝐴𝑀 𝐴𝑀 desloca a economia doméstica na direção de uma transformação estrutural com maior participação de serviços e o quanto isso aparece no equilíbrio depende também de quão facilmente manufaturados e serviços se substituem valor de σ C A desindustrialização pode ser entendida como a queda da participação da manufatura no produto e sobretudo no emprego isto é uma queda de 𝑠𝑀 em que 𝑠𝑀 representa a parcela do emprego ou do valor adicionado na manufatura Em muitas trajetórias históricas 𝑠𝑀 primeiro sobe com o aumento de renda per capita e depois cai quando a economia fica rica e os serviços ganham espaço isso pode ser representado de forma simplificada por uma relação em formato de U invertido entre 𝑠𝑀 e a renda y A desindustrialização prematura ocorre quando esse pico de 𝑠𝑀 acontece em um nível de renda y relativamente baixo e a queda começa antes de o país consolidar capacidades industriais e produtividade elevadas Em termos de trajetória significa que o máximo de 𝑠𝑀𝑦 é atingido cedo e a economia passa a perder indústria quando ainda não alcançou o patamar de renda típico das economias que se industrializaram plenamente Rodrik 2016 e outros autores argumentam que a abertura comercial dos anos 1990 teria interagido com tendências globais e características regionais para gerar efeitos distintos na América Latina e na Ásia Um canal importante é a competição externa em manufaturas e a especialização induzida por vantagens comparativas e por preços relativos internacionais Se o país passa a importar manufaturados a preços relativamente menores eou enfrenta concorrência muito forte de grandes exportadores a demanda por manufaturas domésticas e a lucratividade do setor podem cair deslocando trabalho para serviços eou setores primários reduzindo 𝑠𝑀 Na Ásia a abertura foi frequentemente associada a uma estratégia de crescimento liderado por exportações de manufaturados e integração em cadeias globais de valor Com isso a demanda externa por manufaturas domésticas cresceu e a manufatura pôde absorver trabalho aprender elevar produtividade e diversificar a estrutura produtiva Nesse caso a abertura tende a deslocar a economia na direção de maior produção industrial ao menos por um período relevante sustentando 𝑠𝑀 por mais tempo e postergando a queda Na América Latina a abertura dos anos 1990 combinada em vários casos com apreciação cambial e posteriormente com ciclos de commodities frequentemente reforçou a especialização em bens primários e ampliou a penetração de importações manufatureiras Assim mesmo com ganhos de eficiência parte da indústria local perdeu participação relativa antes de atingir grande escala e sofisticação tecnológica enquanto serviços inclusive serviços de baixa produtividade absorveram parcela crescente do emprego Esse conjunto de forças é compatível com uma queda de 𝑠𝑀 em níveis de renda relativamente mais baixos caracterizando a desindustrialização prematura Além disso mesmo quando a produção manufatureira não cai tanto o progresso técnico tende a reduzir a intensidade de trabalho na manufatura de modo que o emprego industrial pode cair mais cedo do que o produto industrial Isso acelera a queda de 𝑠𝑀 do emprego reforçando o diagnóstico de prematuridade em economias que abrem cedo e enfrentam competição global intensa sem uma trajetória longa de expansão industrial