AI智能总结
What is needed forconvergence? The role of by Bryan Hardy and Can SeverMonetary and Economic Department August 2025 Keywords: productivity, convergence, financialdevelopment, capital, human capital, structural BISWorking Papers are written by members of the Monetary and EconomicDepartment of the Bank for International Settlements, and from time to time by othereconomists, and are published by the Bank. The papers are on subjects of topical This publication is available on the BIS website (www.bis.org). What is Needed for Convergence? The Role of Capital and Finance∗ Bank for International Settlements Abstract:What is needed for poor countries to catch up with rich ones? This paper first documents therole of human capital, physical capital, and financial development in convergence in manufacturinglabor productivity across countries, and then examines the influence of economic structure and financialdevelopment at the aggregate level. Using industry-level data from manufacturing industries in a largeset of countries over the period 1980-2022, we show that manufacturing industries exhibit strong Keywords:Productivity,convergence,financial development,capital,human capital,structuraltransformation 1. Introduction Economists have long sought to understand what makes countries poor or rich, and how poorcountries can grow. Out of this analysis sprang an important prediction: Poor countries have high returnsto investment, and so if capital is allocated efficiently across countries, we should see poor countriesgrow faster than rich ones (Solow 1956). This “unconditional convergence” – whereby poor, low capitalcountries with high returns to capital grow faster than rich ones, capital abundant countries with lowreturns to additional capital – was the subject of considerable debate thereafter. Evidence came in for Evidenceon unconditional convergence in manufacturing industries was more recentlydocumented by Rodrik (2013) and Bénétrix et al. (2015). Rodrik argues that the lack of convergence innon-manufacturing sectors suggests that convergence occurs in only the “modern” parts of the Theresurgence of aggregate convergence suggests that underlying factors that fosterconvergencemight be changing.However,it is still not clear what drives convergence in themanufacturing sector and what the key ingredients are to spur the convergence process and how thesemight contribute to aggregate convergence. Easterly and Levine (2001) suggest that productivity is thekey separator for growth, not factor accumulation. Good quality institutions seem to be a precondition(Acemoglu et al. 2005).1Further, the theoretical literature – in addition to describing the role of capital Thispaper empirically examines the role of(physical and human)capital and financialdevelopment in the convergence process. It first focuses on labor productivity in manufacturing Our analysis first provides evidence on unconditional convergence of labor productivity acrosscountries and 2-digit manufacturing industries, consistent with the findings by Rodrik (2013). Theestimated rate of convergence in our sample suggests that a 2-digit manufacturing industry which isinitially the 25thpercentile of the labor productivity distribution across the sample (a relatively low The analysis then turns to differences inthis convergence based on variation across humancapital intensity (HCI)and physical capital intensity (PCI) of production across 2-digit manufacturingindustries.It finds that industries that rely more on human capital are the ones driving unconditionalconvergence of labor productivity, i.e., industries with higher human capital intensity see fasterconvergence. The findings show that, for instance, an industry at the 25thpercentile of HCI (woodproducts) converges with a rate of 1.1 percent, whereas an industry the 75th percentile of HCI differential convergence boost to the growth rate, based on the differences in HCI across these twoindustries, is 0.8 percentage points. In contrast, physical capital intensity does not correlate with thepace of convergence. However, greater financial development speeds up the convergence process for Finally,we examine country-level data and explore how these results relate to overallmacroeconomic convergence. We assume that as the economies shift from traditional activities (i.e.,agriculture) to more modern production (i.e., broad sectors encompassing industrial production andservices) in the process of structural transformation, they rely more on human capital.2If this is the case,and if the patterns observed at the industry-level hold at the macro-level, one can expect that per capitaGDP convergence should be faster for countries and periods with a higher share of non-agriculturalactivities in GDP. The analysis of per capita GDP shows that unconditional convergence is faster whenthe economy is composed more of human capital-intensive sectors, i.e., industrial production and We follow the empirical framework pro