Forging Ahead, Falling Behind and Fighting Back Read online

  Forging Ahead, Falling Behind and Fighting Back

  To what extent has the British economy declined compared to its competitors and what are the underlying reasons for this decline? Nicholas Crafts, one of the world’s foremost economic historians, tackles these questions in a major new account of Britain’s long-run economic performance. He argues that history matters in interpreting current economic performance, because the present is always conditioned by what went before. Bringing together ideas from economic growth theory and varieties of capitalism to endogenous growth and cliometrics, he reveals the microeconomic foundations of Britain’s economic performance in terms of the impact of institutional arrangements and policy choices on productivity performance. The book traces Britain’s path from the First Industrial Revolution and global economic primacy through its subsequent long-term decline, the strengths and weaknesses of the Thatcherite response and the improvement in relative economic performance that was sustained to the eve of the financial crisis.

  Nicholas Crafts is Professor of Economic History at the University of Warwick. His many publications include The Great Depression of the 1930s: Lessons for Today (2013), co-edited with Peter Fearon, Work and Pay in Twentieth Century Britain (2007), co-edited with Ian Gazeley and Andrew Newell, and British Economic Growth During the Industrial Revolution (1985).

  Forging Ahead, Falling Behind and Fighting Back

  British Economic Growth from the Industrial Revolution to the Financial Crisis

  Nicholas Crafts

  University of Warwick

  University Printing House, Cambridge CB2 8BS, United Kingdom

  One Liberty Plaza, 20th Floor, New York, NY 10006, USA

  477 Williamstown Road, Port Melbourne, VIC 3207, Australia

  314–321, 3rd Floor, Plot 3, Splendor Forum, Jasola District Centre, New Delhi – 110025, India

  79 Anson Road, #06–04/06, Singapore 079906

  Cambridge University Press is part of the University of Cambridge.

  It furthers the University’s mission by disseminating knowledge in the pursuit of education, learning and research at the highest international levels of excellence.

  www.cambridge.org

  Information on this title: www.cambridge.org/9781108424400

  DOI: 10.1017/9781108334907

  © Nicholas Crafts 2018

  This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press.

  First published 2018

  Printed in the United Kingdom by TJ International Ltd. Padstow, Cornwall

  A catalogue record for this publication is available from the British Library.

  ISBN 978-1-108-42440-0 Hardback

  ISBN 978-1-108-43816-2 Paperback

  Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

  Contents

  List of Figures and Tables

  Acknowledgements

  1Introduction

  2The First Industrial Revolution

  3American Overtaking

  4The Interwar Years: Onwards and Downwards

  5Falling Behind in the ‘Golden Age’

  6From the Golden Age to the Financial Crisis

  7Concluding Comments

  References

  Index

  Figures and Tables

  Figures

  1.1Endogenous growth

  Tables

  1.1Real GDP/person (UK = 100 in each year)

  1.2Growth rates of real GDP, population and real GDP/person (% per year)

  2.1Real GDP/person, 1086–1850 ($1990GK)

  2.2Shares of world industrial production (%)

  2.3Sectoral shares in employment (%)

  2.4Labour productivity growth, 1700–1851 (% per year)

  2.5Growth accounting estimates (% per year)

  2.6Contributions to labour productivity growth, 1780–1860 (% per year)

  2.7Steam’s contribution to British labour productivity growth, 1760–1910 (% per year)

  2.8Aspects of broad capital accumulation, 1801–1831 (%)

  2.9GPTs: contributions to labour productivity growth (% per year)

  2.10Leading positive items in current account of balance of payments, 1870

  2.11Industrial shares of employment (%)

  3.1Real GDP/person ($1990GK)

  3.2Sectoral labour productivity growth before the First World War (% per year)

  3.3Contributions to labour productivity growth (% per year)

  3.4Growth of real GDP and TFP, 1856–1937 (% per year)

  3.5Steam power growth and British industrial output and labour productivity growth (% per year)

  3.6The environment for endogenous innovation

  3.7USA/UK productivity levels in 1911 (UK = 100)

  3.8Revealed comparative advantage rankings

  3.9Impact of changing sectoral weights on labour productivity growth in US manufacturing, 1899–1909 (% per year)

  4.1Real GDP/person ($GK1990)

  4.2Contributions to labour productivity growth (% per year)

  4.3Crude TFP growth in major sectors (% per year)

  4.4Investments in broad capital

  4.5Unemployment rates (%)

  4.6Sectoral contributions to manufacturing labour productivity growth (%)

  4.7Real output/hour worked in manufacturing

  5.1Real GDP/person ($GK1990)

  5.2Investment in broad capital, 1970

  5.3Contributions to labour productivity gap (percentage points)

  5.4Contributions to labour productivity growth, 1950–1973 (% per year)

  5.5Crude TFP growth in major sectors, 1950–1973 (% per year)

  6.1Real GDP per person

  6.2Rates of growth of real GDP/person and real GDP/hour worked (% per year)

  6.3Investment in broad capital, c. 2000

  6.4Contributions to labour productivity gap (percentage points)

  6.5Contributions to labour productivity growth, 1973–2007 (% per year)

  6.6Levels of productivity (UK = 100 in each year)

  6.7Labour productivity growth in the market sector, 1995–2007 (% per year)

  Acknowledgements

  This book has been developed from the Ellen McArthur Lectures which I gave at the University of Cambridge in 2009. I was honoured by the invitation to give the lectures. The opportunity encouraged me further to explore themes in my research which had not previously been thought through, especially with regard to the implication of Britain’s early start as an industrial nation.

  When I presented the lectures I received excellent support from Martin Daunton and Leigh Shaw-Taylor. Martin Daunton and Tim Leunig read a draft of the book and made valuable suggestions. Michael Watson at Cambridge University Press has been supportive throughout and has been astonishingly patient with my slow progress in delivering the manuscript. I am grateful to all of them.

  1

  Introduction

  This book examines the British economy’s growth performance over the last 250 years. The focal point is to offer an interpretation – informed by ideas from growth economics, and firmly grounded in empirical evidence – of the relative economic decline that characterized the period from the mid-nineteenth century, when Britain had the highest per capita income of any major economy, to the early 1980s, when this had fallen below the West-European average. T
his will entail an analysis of the experience of economic growth from the Industrial Revolution to the eve of the financial crisis which erupted in 2007.

  The concept of ‘relative economic decline’ relates to international comparisons of the level of real Gross Domestic Product (GDP) per person. As applied to Britain, it means that over many decades economic growth was slower than in a peer group of other countries, with the result that they first caught up, and then overtook, British income levels. As is reported in Table 1.1, this describes the economic history of the post-Industrial Revolution period through the 1970s. Relative economic decline was most apparent vis-à-vis the United States, from the American Civil War to 1950 and, compared with European countries, during the 1950s to the 1970s.

  Table 1.1 Real GDP/person (UK = 100 in each year)

  USA Germany France

  1820 65.6 51.9 54.7

  1870 76.6 57.6 58.8

  1913 107.7 74.1 70.8

  1929 125.3 73.6 85.6

  1937 103.4 75.3 72.2

  1950 137.8 61.7 74.7

  1979 142.7 115.9 111.1

  2007 132.9 107.0 98.6

  Notes: Estimates refer to West Germany in 1950 and 1979. Purchasing power parity estimates in $1990GK for 1870–1979 and in $2015EKS for 2007.

  Sources: Maddison (2010) and The Conference Board (2016).

  Relative economic decline did not mean that British economic growth slowed down. On the contrary, as is shown in Table 1.2, the long-run tendency was for the rate of growth of real GDP per person to increase over time. The acceleration in economic growth which Britain experienced as result of the Industrial Revolution represents the transition to ‘modern economic growth’ (Kuznets, 1966) where technological progress took centre stage. From the Industrial Revolution to the First World War, growth averaged a little under 1 per cent per year, roughly double the rate from 1650 to 1780 – itself well above the 0.2 per cent average over the previous 400 years – but less than half that achieved since the Second World War. The problem was rather that growth in other countries increased by more than in Britain as faster technological advance became possible.

  Table 1.2 Growth rates of real GDP, population and real GDP/person (% per year)

  GDP Population Real GDP /person

  1500–1650 0.59 0.60 –0.01

  1650–1780 0.71 0.24 0.47

  1780–1820 1.43 1.22 0.21

  1820–1870 2.12 1.24 0.88

  1870–1913 1.90 0.89 1.01

  1929–1937 1.99 0.44 1.55

  1950–1979 2.63 0.40 2.23

  1979–2007 2.54 0.32 2.22

  Note: Estimates based on England up to 1700, Britain 1700–1870, United Kingdom 1870–2007.

  Sources: Broadberry et al. (2015) and The Maddison Project database.

  Evidently, growth comparisons, whether inter-temporal or international, need to be handled with care. It is important to take into account what is feasible, and to recognize that relative economic decline does not always connote ‘failure’. It seems clear that the accumulation of knowledge and human capital characteristic of the last 100 years has been conducive to faster technological progress in the advanced economies, as is reflected in their capacity to exploit major new technologies increasingly quickly (Crafts, 2012). Growth of real GDP per person of around 2 per cent per year was not feasible in 1800 but quite normal 200 years later. Similarly, growth possibilities may vary across countries at a point in time because of different scope for catch-up or the ‘inappropriateness’ of technological change.

  The former is widely recognized and with the availability of purchasing power parity adjusted series for relative income levels can now be taken properly into account. Countries grow faster when they embark on catch-up from an initially low income and productivity level. No Western European country could expect to grow at a double-digit pace as China has in the recent past. Equally, Britain as the first industrial nation, could expect to be caught up as modern economic growth spread – reflected in relative economic decline compared with European countries in the nineteenth century. On the other hand, being overtaken by its European peer group, as happened to Britain in the 1960s and 1970s, surely is a diagnostic of a growth failure since there is no reason to think that other countries had access to superior technology or a more favourable geography.

  Adoption of a new technology is not always appropriate – it may be profitable in some countries but not others because cost or demand conditions differ. It follows that different technological choices may be rational and the technological playing field may not be level. The appropriateness of technology may be affected by relative factor prices perhaps differing on account of geography or the level of development. It is widely remarked that this is an important issue in the viability of technologies developed by advanced economies for adoption in poor developing countries (Allen, 2012). But, in past times, appropriateness was relevant to the diffusion of technology between leading economies both with regard to other countries’ ability to emulate Britain at the time of the Industrial Revolution, and in terms of American technology’s suitability for adoption in Europe at the time of the ‘second Industrial Revolution’ a hundred years later.

  Growth economics now offers valuable analytical tools with which to develop an explanation for relative economic decline which was not really the case when the traditional neoclassical economic growth model ruled the roost. This viewed the sources of economic growth as growth in the capital stock and the labour force, and improvements in technology which raised the productivity of these inputs. This model has two key assumptions, namely, that capital accumulation is subject to diminishing returns and that technological progress is exogenous and universally available. These assumptions are fundamental to two well-known predictions of the model about the long run, namely, that increasing the rate of investment has no effect on the steady-state rate of economic growth and that all countries converge to the same income level as initially backward countries automatically enjoy rapid catch-up growth.1

  Although some insights from this model have found favour (and an empirical technique derived from it, growth accounting, has been widely used in economic history) it is fair to say that the pure neoclassical model has been regarded by most economic historians, as unhelpful much of the time. In particular, the notions of universal technology and long-run income convergence have seemed far-fetched to scholars accustomed to thinking in terms of, say, the new institutional economic history with its emphasis on the importance of institutions and political economy considerations to growth outcomes. Moreover, this model cannot really cope with the leading economy being overtaken and, after all, this is at the heart of Britain’s relative economic decline.

  The so-called ‘new’ growth economics offers models with more attractive features. These include acceptance that institutions and policy can affect the growth rate, and can promote divergence in growth outcomes and, associated with this, the recognition that catching-up is not automatic. The most useful of these new models embody the idea of endogenous innovation; they consider that technological advance, whether through invention or diffusion, is influenced by economic incentives, in particular, expected profitability and they drop the assumption that technology is universal. Technologies are developed to address market demands in particular locations and may not be appropriate elsewhere (Acemoglu, 1998). Carefully deployed, these ideas can inform an appraisal of controversies surrounding British growth performance.

  Broadly speaking, new growth economics suggests that there are two important aspects of the incentive structures that influence the decisions to invest and to innovate which matter for growth outcomes, namely, their impact on expected returns and on agency problems (Aghion and Howitt, 1998). Thus, institutions and policies that reduce the supply price of capital or research inputs, or reduce fears of expropriation, can increase innovative effort, speed up technology transfer and enhance the chances of rapid catch-up growth. Innovative effort is also positively affected by greater market size, which
makes it easier to cover the fixed costs of innovating. Since effective and timely adoption of new technologies tends to be costly to the management of firms in terms of the effort required, it is also important that managers are incentivized to work hard on behalf of the owners – when this is not the case we speak of performance being jeopardized by principal–agent problems. Unless there are large external shareholders who can internalize the benefits of effective control of management, strong (though less than perfect) competition tends to be important in underpinning TFP growth (Nickell, 1996).

  These ideas also resonate with economic historians’ discussions of the international diffusion of technology. In particular, there is an obvious connection with the idea of ‘social capability’ used by Abramovitz and David (1996). But it should also be noted that these authors also stress the importance of ‘technological congruence’ in catching up or falling behind. Here the point is that the cost-effectiveness of a technology may vary across countries where demand or cost conditions are different. An interesting aspect of this, as pointed out by Abramovitz (1986) is that social capability is not an absolute but may vary according to the technology in question – for example, institutions and policies which were excellent for the diffusion of Fordist production techniques in manufacturing in the 1950s, may not be ideal to facilitate rapid uptake of ICT in services in the 1990s.

  The key ideas are captured in Figure 1.1, which is adapted from Carlin and Soskice (2006). In this figure x is the rate of (labour-augmenting) technological progress and ǩ is the capital to effective labour ratio. The upward-sloping (Schumpeter) line reflects the endogeneity of technological progress based on the assumption a larger market increases innovative effort because it is potentially more profitable, since success will be rewarded by greater sales. With more capital per unit of effective labour there will be higher income per person so the Schumpeter line is upward-sloping. The downward-sloping (Solow) line represents points which are consistent with the steady-state relationship between technological progress and capital per effective unit of labour. The steady-state is characterized by balanced growth in which the capital stock grows at the same rate as the sum of labour force growth and the rate of technological progress. When this is the case the capital to output ratio is constant and so is the ratio of capital to an effective unit of labour. For a given savings rate, the growth of the capital stock is faster the lower the capital to output ratio. With a ‘well-behaved’ production function, lower capital per effective unit of labour means a lower capital to output ratio. Thus, the Solow line will be downward sloping. The equilibrium rate of technological progress is established by the intersection of these two lines.