Studies of firm-level data have shown that there is a huge dispersion of productivity across firm... more Studies of firm-level data have shown that there is a huge dispersion of productivity across firms even when industries are narrowly defined. So there is a significant opportunity for the least productive firms to catch up to the most productive. The formers' convergence could therefore constitute an important part of productivity growth at the macroeconomic level. This article sheds light on this convergence process in the 1990s and the 2000s in France and on some of the factors which can explain it. Productivity convergence was stronger for labour productivity than for total factor productivity. But most importantly the speed of convergence has slowed during the course of the 1990s, a fact which is explained principally by the acceleration of the productivity of firms on the technological frontier. Three possible explanations of these stylised facts are considered: globalisation, information technology, and competition. Globalisation and information technology may have benefited the most productive firms more and the growth of competition may at the same time have stimulated the productivity of firms at the frontier while discouraging the convergence of the least productive firms.
Introduction 2 The relationship between ISTC and TFP 3 Jorgensonian growth accounting 4 Should th... more Introduction 2 The relationship between ISTC and TFP 3 Jorgensonian growth accounting 4 Should the whole of output be deflated by the price of consumption? 5 What proportion of growth is due to embodiment (ISTC)?
We argue that unmeasured investments in intangible organizational capital-associated with the rol... more We argue that unmeasured investments in intangible organizational capital-associated with the role of information and communications technology (ICT) as a 'general purpose technology'-can explain the divergent U.S. and U.K. TFP performance after 1995. GPT stories suggest that measured TFP should rise in ICT-using sectors, perhaps with long lags. Contemporaneously, investments in ICT may in fact be associated with lower TFP as resources are diverted to reorganization and learning. In both the U.S. and U.K., we find a strong correlation between ICT use and industry TFP growth. The U.S. results, in particular, are consistent with GPT stories: the TFP acceleration was located primarily in ICT-using industries and is positively correlated with industry ICT capital growth from the 1980s and early 1990s. Indeed, as GPT stories suggest, controlling for past ICT growth, industry TFP growth appears negatively correlated with increases in ICT capital services in the late 1990s. A somewhat different picture emerges for the U.K. TFP growth does not appear correlated with lagged ICT capital growth. But TFP growth in the late 1990s is strongly and positively associated with the growth of ICT capital services, while being strongly and negatively associated with the growth of ICT investment.
The purpose of this chapters is to estimate the level of protection accorded to British industry ... more The purpose of this chapters is to estimate the level of protection accorded to British industry on the eve of entry into the European Community, concentrating particularly on tariff protection for manufactured goods. As the theory of effective protection provides the method by which the problem will be attacked, the second section therefore discusses and develops the theory before we turn, in the third section, to considering the available data and the problems of estimation.
What effect, if any, do changes in the terms of trade have on the level of output (GDP) or welfar... more What effect, if any, do changes in the terms of trade have on the level of output (GDP) or welfare? I examine this issue through two versions of a textbook, Heckscher‐Ohlin‐Samuelson (HOS), two‐good model of a small, open economy. In the first version both goods are for final consumption. In the second, one good is an imported intermediate input into the other. In both versions, economic theory suggests that an improvement in the terms of trade raises welfare (consumption) but leaves aggregate output (GDP) unchanged. I then show that a national income accountant applying the principles of the 2008 System of National Accounts (SNA) would reach the same conclusions. This follows from a continuous‐time analysis using Divisia index numbers. However in the case where imports are intermediate inputs and competition is imperfect, an improvement in the terms of trade does raise GDP: the size of the effect depends on the size of the markup of price over marginal revenue. I argue that the continuous time Divisia approach is the right framework for national income accounting, even though it can only be implemented approximately in practice. If the aim is to find the best approximation to the Divisia index, then the chained Fisher index (as used in the US and Canadian national accounts) or the chained Törnqvist are better approximations than is the chained Laspeyres (as used in Europe).
I propose a new explanation for the UK productivity puzzle. I graft the Lewis (1954) model onto a... more I propose a new explanation for the UK productivity puzzle. I graft the Lewis (1954) model onto a standard Solow growth model. What I call the neo-Lewis model is identical to the Solow model in good times. But in bad times foreign demand for a country's exports is constrained below potential supply. This makes labour productivity growth depend negatively on the growth of labour input. I also argue that the neo-Lewis model can explain the fall in TFP growth, in the UK and elsewhere, after 2007. The predictions of the neo-Lewis model are tested on data for 23 advanced countries and also on a larger sample of 52 countries and find support.
I set out a general algorithm for calculating true cost-of-living indices when demand is not homo... more I set out a general algorithm for calculating true cost-of-living indices when demand is not homothetic and when the number of products may be large. The non-homothetic case is the important one empirically as Engel's Law demonstrates. The algorithm can be applied in both time series and cross section, eg to estimate true PPPs. The algorithm can also be used to estimate true producer price indices and consequently TFP in the presence of input-biased economies of scale. The data required are no more than are needed for the construction of conventional index numbers.
How big a boost to long run growth can countries expect from the ICT revolution? I use the result... more How big a boost to long run growth can countries expect from the ICT revolution? I use the results of growth accounting and the insights from a two-sector growth model to answer this question. The use of a two-sector rather than a one-sector model is required because of the very rapid rate at which the prices of ICT products have fallen in the past and are expected to fall in the future. According to the two-sector model, the main boost to growth comes from ICT use, not ICT production. Even a country which has zero ICT production can benefit via improving terms of trade. In the long run, the falling relative price of ICT products boosts the growth of GDP and consumption by inducing faster accumulation of ICT capital. I quantify this effect on the long run growth rate of 15 European and 4 non-European countries, using data from the EU KLEMS database. The ICT intensity of production (the ICT income share) is much lower in many European countries than it is in the United States or Sweden. Nevertheless the contribution to the long run growth of labour productivity stemming from even the current levels of ICT intensity is substantial: about half a percent per annum on average in the countries studied here. Eventually, the ICT revolution may diffuse more widely so ICT intensity may reach at least the same level as currently in the U.S. or Sweden, which would add a further 0.2 percentage points per annum to long run growth.
Should we use ex post or ex ante measures of user costs to calculate the contribution of capital ... more Should we use ex post or ex ante measures of user costs to calculate the contribution of capital in a growth accounting exercise? The answer, based on a simple model of temporary equilibrium, is that ex post is better in theory. In practice researchers usually calculate ex post user costs by assuming that the rate of return is equalized across assets. But this is only true if expectations are correct. In general, the ex post rate of return differs between assets, even though ex ante it is the same. I propose a hybrid method. The index of capital services is estimated using ex ante weights; the contribution of capital is the growth of this index multiplied by the ex post income share of capital. I show that this method is theoretically correct if the production function is CES. I compare the ex post, ex ante and hybrid methods using data for 31 U.K. industries from 1970 to 2000. Note: I am grateful to Dale Jorgenson, to two anonymous referees, and to the editor, Bart van Ark, for helpful comments. This paper benefited from discussion at the 2nd consortium meeting of the EU KLEMS group in Helsinki, June 9-11, 2005. I owe thanks to the Bank of England for permission to use the Bank of England Industry Dataset. This research was made possible by financial support under the 6th Framework Programme of the European Commission to the "EU KLEMS Project on Productivity in the European Union." The views expressed are my own.
We argue that unmeasured investments in intangible organizational capital-associated with the rol... more We argue that unmeasured investments in intangible organizational capital-associated with the role of information and communications technology (ICT) as a 'general purpose technology'-can explain the divergent U.S. and U.K. TFP performance after 1995. GPT stories suggest that measured TFP should rise in ICT-using sectors, perhaps with long lags. Contemporaneously, investments in ICT may in fact be associated with lower TFP as resources are diverted to reorganization and learning. In both the U.S. and U.K., we find a strong correlation between ICT use and industry TFP growth. The U.S. results, in particular, are consistent with GPT stories: the TFP acceleration was located primarily in ICT-using industries and is positively correlated with industry ICT capital growth from the 1980s and early 1990s. Indeed, as GPT stories suggest, controlling for past ICT growth, industry TFP growth appears negatively correlated with increases in ICT capital services in the late 1990s. A somewhat different picture emerges for the U.K. TFP growth does not appear correlated with lagged ICT capital growth. But TFP growth in the late 1990s is strongly and positively associated with the growth of ICT capital services, while being strongly and negatively associated with the growth of ICT investment.
1. Introduction 3 2. The ARD 5 3. Structural change in UK manufacturing, 1979-89 6 4. Decomposing... more 1. Introduction 3 2. The ARD 5 3. Structural change in UK manufacturing, 1979-89 6 4. Decomposing productivity growth 8 5. Decomposing productivity growth in practice, 1979-89 5.1 The role of closures 5.2 Productivity growth amongst survivors: downsizing versus upsizing
We examine the hypothesis that capacity can be permanently damaged by financial, particularly ban... more We examine the hypothesis that capacity can be permanently damaged by financial, particularly banking, crises. A model which allows a financial crisis to have both a short-run effect on the growth rate of labour productivity and a long-run effect on its level is estimated on 61 countries over 1954-2010. A banking crisis as defined by Reinhart and Rogoff reduces the long-run level of GDP per worker, and also that of capital per worker, by on average 1.1%, for each year that the crisis lasts; it also reduces the TFP level by 0.8%. The long run, negative effect on the level of GDP per capita, 1.8%, is substantially larger. So there is also a hit to employment. The effects on labour productivity, capital and TFP are larger in developing than in developed countries; the opposite is the case for employment.
Introduction 2 The relationship between ISTC and TFP 3 Jorgensonian growth accounting 4 Should th... more Introduction 2 The relationship between ISTC and TFP 3 Jorgensonian growth accounting 4 Should the whole of output be deflated by the price of consumption? 5 What proportion of growth is due to embodiment (ISTC)?
Measured by the growth of productivity, the performance of the UK economy was excellent in the pe... more Measured by the growth of productivity, the performance of the UK economy was excellent in the period 1990-2007. Performance on other dimensions such as employment, unemployment and inflation was also good. But the financial crisis and the Great Recession which began in Spring 2008 have dealt this optimistic picture a devastating blow. Both GDP and GDP per hour have fallen and are still below the level reached at the peak of the boom. In this paper I first analyse why one might have expected prior to the crisis that the UK's excellent growth would have continued into the future. I employ a two-sector model to produce a projection of the future growth rate of GDP per hour in the market sector, based entirely on pre-crisis data: the resulting number is 2.61% p.a. I then consider what has happened since 2008 and discuss a wide range of hypotheses which seek to explain the productivity collapse, including the impact of austerity. Most of the conclusions here are negative: the explanation in question doesn't work. I next turn to the long run impact of financial crises, particularly banking crises, on productivity, capital, TFP and employment. Based on a crosscountry panel analysis of 61 countries over 1950-2010, I argue that banking crises generally have a long run impact on the level of productivity but not necessarily on its long run growth rate. I therefore predict that the UK will eventually return to the growth rate predicted prior to the crisis. This prediction is however conditional on the UK continuing to follow good policies in other respects, in particular not allowing the government debt-GDP ratio to rise excessively. The permanent reduction in the level of GDP per worker resulting from the crisis could however be substantial, about 4-5%. The crosscountry evidence also suggests that there are permanent effects on employment. This suggests an even larger hit to the level of GDP per capita, about 7-9%.
Studies of firm-level data have shown that there is a huge dispersion of productivity across firm... more Studies of firm-level data have shown that there is a huge dispersion of productivity across firms even when industries are narrowly defined. So there is a significant opportunity for the least productive firms to catch up to the most productive. The formers' convergence could therefore constitute an important part of productivity growth at the macroeconomic level. This article sheds light on this convergence process in the 1990s and the 2000s in France and on some of the factors which can explain it. Productivity convergence was stronger for labour productivity than for total factor productivity. But most importantly the speed of convergence has slowed during the course of the 1990s, a fact which is explained principally by the acceleration of the productivity of firms on the technological frontier. Three possible explanations of these stylised facts are considered: globalisation, information technology, and competition. Globalisation and information technology may have benefited the most productive firms more and the growth of competition may at the same time have stimulated the productivity of firms at the frontier while discouraging the convergence of the least productive firms.
Introduction 2 The relationship between ISTC and TFP 3 Jorgensonian growth accounting 4 Should th... more Introduction 2 The relationship between ISTC and TFP 3 Jorgensonian growth accounting 4 Should the whole of output be deflated by the price of consumption? 5 What proportion of growth is due to embodiment (ISTC)?
We argue that unmeasured investments in intangible organizational capital-associated with the rol... more We argue that unmeasured investments in intangible organizational capital-associated with the role of information and communications technology (ICT) as a 'general purpose technology'-can explain the divergent U.S. and U.K. TFP performance after 1995. GPT stories suggest that measured TFP should rise in ICT-using sectors, perhaps with long lags. Contemporaneously, investments in ICT may in fact be associated with lower TFP as resources are diverted to reorganization and learning. In both the U.S. and U.K., we find a strong correlation between ICT use and industry TFP growth. The U.S. results, in particular, are consistent with GPT stories: the TFP acceleration was located primarily in ICT-using industries and is positively correlated with industry ICT capital growth from the 1980s and early 1990s. Indeed, as GPT stories suggest, controlling for past ICT growth, industry TFP growth appears negatively correlated with increases in ICT capital services in the late 1990s. A somewhat different picture emerges for the U.K. TFP growth does not appear correlated with lagged ICT capital growth. But TFP growth in the late 1990s is strongly and positively associated with the growth of ICT capital services, while being strongly and negatively associated with the growth of ICT investment.
The purpose of this chapters is to estimate the level of protection accorded to British industry ... more The purpose of this chapters is to estimate the level of protection accorded to British industry on the eve of entry into the European Community, concentrating particularly on tariff protection for manufactured goods. As the theory of effective protection provides the method by which the problem will be attacked, the second section therefore discusses and develops the theory before we turn, in the third section, to considering the available data and the problems of estimation.
What effect, if any, do changes in the terms of trade have on the level of output (GDP) or welfar... more What effect, if any, do changes in the terms of trade have on the level of output (GDP) or welfare? I examine this issue through two versions of a textbook, Heckscher‐Ohlin‐Samuelson (HOS), two‐good model of a small, open economy. In the first version both goods are for final consumption. In the second, one good is an imported intermediate input into the other. In both versions, economic theory suggests that an improvement in the terms of trade raises welfare (consumption) but leaves aggregate output (GDP) unchanged. I then show that a national income accountant applying the principles of the 2008 System of National Accounts (SNA) would reach the same conclusions. This follows from a continuous‐time analysis using Divisia index numbers. However in the case where imports are intermediate inputs and competition is imperfect, an improvement in the terms of trade does raise GDP: the size of the effect depends on the size of the markup of price over marginal revenue. I argue that the continuous time Divisia approach is the right framework for national income accounting, even though it can only be implemented approximately in practice. If the aim is to find the best approximation to the Divisia index, then the chained Fisher index (as used in the US and Canadian national accounts) or the chained Törnqvist are better approximations than is the chained Laspeyres (as used in Europe).
I propose a new explanation for the UK productivity puzzle. I graft the Lewis (1954) model onto a... more I propose a new explanation for the UK productivity puzzle. I graft the Lewis (1954) model onto a standard Solow growth model. What I call the neo-Lewis model is identical to the Solow model in good times. But in bad times foreign demand for a country's exports is constrained below potential supply. This makes labour productivity growth depend negatively on the growth of labour input. I also argue that the neo-Lewis model can explain the fall in TFP growth, in the UK and elsewhere, after 2007. The predictions of the neo-Lewis model are tested on data for 23 advanced countries and also on a larger sample of 52 countries and find support.
I set out a general algorithm for calculating true cost-of-living indices when demand is not homo... more I set out a general algorithm for calculating true cost-of-living indices when demand is not homothetic and when the number of products may be large. The non-homothetic case is the important one empirically as Engel's Law demonstrates. The algorithm can be applied in both time series and cross section, eg to estimate true PPPs. The algorithm can also be used to estimate true producer price indices and consequently TFP in the presence of input-biased economies of scale. The data required are no more than are needed for the construction of conventional index numbers.
How big a boost to long run growth can countries expect from the ICT revolution? I use the result... more How big a boost to long run growth can countries expect from the ICT revolution? I use the results of growth accounting and the insights from a two-sector growth model to answer this question. The use of a two-sector rather than a one-sector model is required because of the very rapid rate at which the prices of ICT products have fallen in the past and are expected to fall in the future. According to the two-sector model, the main boost to growth comes from ICT use, not ICT production. Even a country which has zero ICT production can benefit via improving terms of trade. In the long run, the falling relative price of ICT products boosts the growth of GDP and consumption by inducing faster accumulation of ICT capital. I quantify this effect on the long run growth rate of 15 European and 4 non-European countries, using data from the EU KLEMS database. The ICT intensity of production (the ICT income share) is much lower in many European countries than it is in the United States or Sweden. Nevertheless the contribution to the long run growth of labour productivity stemming from even the current levels of ICT intensity is substantial: about half a percent per annum on average in the countries studied here. Eventually, the ICT revolution may diffuse more widely so ICT intensity may reach at least the same level as currently in the U.S. or Sweden, which would add a further 0.2 percentage points per annum to long run growth.
Should we use ex post or ex ante measures of user costs to calculate the contribution of capital ... more Should we use ex post or ex ante measures of user costs to calculate the contribution of capital in a growth accounting exercise? The answer, based on a simple model of temporary equilibrium, is that ex post is better in theory. In practice researchers usually calculate ex post user costs by assuming that the rate of return is equalized across assets. But this is only true if expectations are correct. In general, the ex post rate of return differs between assets, even though ex ante it is the same. I propose a hybrid method. The index of capital services is estimated using ex ante weights; the contribution of capital is the growth of this index multiplied by the ex post income share of capital. I show that this method is theoretically correct if the production function is CES. I compare the ex post, ex ante and hybrid methods using data for 31 U.K. industries from 1970 to 2000. Note: I am grateful to Dale Jorgenson, to two anonymous referees, and to the editor, Bart van Ark, for helpful comments. This paper benefited from discussion at the 2nd consortium meeting of the EU KLEMS group in Helsinki, June 9-11, 2005. I owe thanks to the Bank of England for permission to use the Bank of England Industry Dataset. This research was made possible by financial support under the 6th Framework Programme of the European Commission to the "EU KLEMS Project on Productivity in the European Union." The views expressed are my own.
We argue that unmeasured investments in intangible organizational capital-associated with the rol... more We argue that unmeasured investments in intangible organizational capital-associated with the role of information and communications technology (ICT) as a 'general purpose technology'-can explain the divergent U.S. and U.K. TFP performance after 1995. GPT stories suggest that measured TFP should rise in ICT-using sectors, perhaps with long lags. Contemporaneously, investments in ICT may in fact be associated with lower TFP as resources are diverted to reorganization and learning. In both the U.S. and U.K., we find a strong correlation between ICT use and industry TFP growth. The U.S. results, in particular, are consistent with GPT stories: the TFP acceleration was located primarily in ICT-using industries and is positively correlated with industry ICT capital growth from the 1980s and early 1990s. Indeed, as GPT stories suggest, controlling for past ICT growth, industry TFP growth appears negatively correlated with increases in ICT capital services in the late 1990s. A somewhat different picture emerges for the U.K. TFP growth does not appear correlated with lagged ICT capital growth. But TFP growth in the late 1990s is strongly and positively associated with the growth of ICT capital services, while being strongly and negatively associated with the growth of ICT investment.
1. Introduction 3 2. The ARD 5 3. Structural change in UK manufacturing, 1979-89 6 4. Decomposing... more 1. Introduction 3 2. The ARD 5 3. Structural change in UK manufacturing, 1979-89 6 4. Decomposing productivity growth 8 5. Decomposing productivity growth in practice, 1979-89 5.1 The role of closures 5.2 Productivity growth amongst survivors: downsizing versus upsizing
We examine the hypothesis that capacity can be permanently damaged by financial, particularly ban... more We examine the hypothesis that capacity can be permanently damaged by financial, particularly banking, crises. A model which allows a financial crisis to have both a short-run effect on the growth rate of labour productivity and a long-run effect on its level is estimated on 61 countries over 1954-2010. A banking crisis as defined by Reinhart and Rogoff reduces the long-run level of GDP per worker, and also that of capital per worker, by on average 1.1%, for each year that the crisis lasts; it also reduces the TFP level by 0.8%. The long run, negative effect on the level of GDP per capita, 1.8%, is substantially larger. So there is also a hit to employment. The effects on labour productivity, capital and TFP are larger in developing than in developed countries; the opposite is the case for employment.
Introduction 2 The relationship between ISTC and TFP 3 Jorgensonian growth accounting 4 Should th... more Introduction 2 The relationship between ISTC and TFP 3 Jorgensonian growth accounting 4 Should the whole of output be deflated by the price of consumption? 5 What proportion of growth is due to embodiment (ISTC)?
Measured by the growth of productivity, the performance of the UK economy was excellent in the pe... more Measured by the growth of productivity, the performance of the UK economy was excellent in the period 1990-2007. Performance on other dimensions such as employment, unemployment and inflation was also good. But the financial crisis and the Great Recession which began in Spring 2008 have dealt this optimistic picture a devastating blow. Both GDP and GDP per hour have fallen and are still below the level reached at the peak of the boom. In this paper I first analyse why one might have expected prior to the crisis that the UK's excellent growth would have continued into the future. I employ a two-sector model to produce a projection of the future growth rate of GDP per hour in the market sector, based entirely on pre-crisis data: the resulting number is 2.61% p.a. I then consider what has happened since 2008 and discuss a wide range of hypotheses which seek to explain the productivity collapse, including the impact of austerity. Most of the conclusions here are negative: the explanation in question doesn't work. I next turn to the long run impact of financial crises, particularly banking crises, on productivity, capital, TFP and employment. Based on a crosscountry panel analysis of 61 countries over 1950-2010, I argue that banking crises generally have a long run impact on the level of productivity but not necessarily on its long run growth rate. I therefore predict that the UK will eventually return to the growth rate predicted prior to the crisis. This prediction is however conditional on the UK continuing to follow good policies in other respects, in particular not allowing the government debt-GDP ratio to rise excessively. The permanent reduction in the level of GDP per worker resulting from the crisis could however be substantial, about 4-5%. The crosscountry evidence also suggests that there are permanent effects on employment. This suggests an even larger hit to the level of GDP per capita, about 7-9%.
Uploads
Papers by Nick Oulton