The Great Economists

by Linda Yueh

  After he received his PhD in economics that year, he joined the Massachusetts Institute of Technology, where he became a professor in 1958. Solow spent his academic career at this leading economics faculty, though he was also visiting professor at Cambridge and Oxford universities in the 1960s.

  Solow was active in public policy from the start. After obtaining his PhD, he took on consulting assignments for the RAND Corporation in 1952. During his time working with the President’s Council of Economic Advisers from 1962–68, Solow helped draft the Keynesian-influenced economic policies that were the hallmark of the John F. Kennedy and Lyndon B. Johnson administrations. In 1965–69, he served on President Johnson’s Committee on Technology, Automation and Economic Progress, and then on President Richard Nixon’s Commission on Income Maintenance from 1969–70. Solow even spent a spell as a central banker when he was director and later chairman of the board of the Federal Reserve Bank of Boston from 1975 to 1980. In recognition of his long public service, Solow was awarded the Presidential Medal of Freedom in 2014, the highest honour granted to civilians by the United States government.

  Solow had received accolades from the start of his career. In 1961 he received the prestigious John Bates Clark Medal, awarded to the best US economist under the age of forty. It is now often viewed as a precursor to the Nobel Prize. He was well regarded throughout his career, which included serving as president of the American Economic Association in 1979. He is also a past president of the Econometric Society, member of the National Science Board, Fellow of the British Academy and recipient of the National Medal of Science. Perhaps unsurprisingly, Solow was awarded the highest prize in economics in 1987 for his work on economic growth. Before the award he had been mentioned regularly as a possible Nobel laureate, which led him to quip: ‘My friends have been telling me that I would get it if I lived long enough.’3

  Yet, economic growth was not his first interest. Solow had intended to focus on statistics and econometrics in his academic career. He attributes his switch to macroeconomics to chance. He was allocated an office next to Nobel laureate Paul Samuelson at MIT. In his 1987 Nobel Prize autobiography, he commented: ‘Thus began what is now almost forty years of almost daily conversations about economics, politics, our children, cabbages and kings.’4

  Solow retired in 1995 to make room for younger scholars, though he remains active in numerous scholarly projects, and still occupies an office that was next to Samuelson’s until the latter’s death in 2009.

  The Solow growth model

  In influential articles in 1956 and 1957,5 Robert Solow laid the foundations for understanding economic growth. The Solow growth model is the standard neoclassical model that is taught in every textbook, including mine. The best-known result from growth models is the Solow residual. The Solow residual refers to the unexplained portion of economic growth which isn’t attributed to adding inputs such as workers and capital. The residual captures technological progress, which generates more output from a set of inputs. Of course, it also captures anything else not related to inputs of labour and investment, so temporary rises in government spending and monetary easing also get included. It means that some, but not all, of what is captured in the Solow residual is the productivity advancing technology needed to sustain economic growth over the longer term. This is the TFP (total factor productivity) mentioned earlier.

  Across countries, there is a clear association between periods of high output growth and significant technological progress. Developed nations all grew well between 1950 and 1973, and then slowed together during 1974–87. There seems to be a connection with the adoption of similar technologies. For instance, the strong period of growth in the 1950s and 60s is associated with post-war technological advances, such as widespread air travel and industrial robots.

  Curiously, recent technological improvements, centred on computing, information and communication technologies (ICT) and the internet, do not seem to have raised productivity across the economy. Solow’s 1987 observation that ‘You can see the computer age everywhere but in the productivity statistics’ is known as the Solow paradox.6 He revisited this question decades later, but concluded that we still do not know, as the role of computing is still evolving. Solow points out that since our lives and work have been transformed by computers, this technology should have improved our productivity. But productivity growth was slow from around 1970 to 1995, which is the period when computing took off. In a shorter period, from 1995 to 2000, productivity growth was faster, which may be attributed to the lagging effects of adopting computing. Solow believes it takes time for businesses to learn to use computers productively, so the early years were not a good indicator. In a 2002 interview he doubted that productivity growth would revert to the fast pace seen previously because ‘Comparing the computer with electricity or the internal combustion engine just doesn’t seem to me to be justified yet.’ Solow also revealed: ‘I always thought that the main difference the computer made in my office was that before the computer my secretary used to work for me, and afterward I worked for my secretary.’7

  Solow’s scepticism reflects one view that the ICT revolution would not generate as much economy-wide productivity improvement as the earlier Industrial Revolution that introduced general purpose technologies such as the steam engine during Adam Smith’s era, or the Second Industrial Revolution that saw the introduction of railways and electrification during the period lasting from the late nineteenth century until the First World War. Others disagree and expect that productivity will improve once these new ICT and digital technologies become truly embedded into work practices and businesses. A major challenge to Solow’s view is related to technology. The developers of endogenous growth models from the 1960s onwards criticized Solow for not explaining where technology came from.

  Endogenous growth models treat technology as determined within the model; in other words, ‘endogenously’ generated by the capital and labour within an economy. The neoclassical Solow model was alleged to treat technological progress as if it were ‘manna from heaven’. By contrast, endogenous growth theories attempt to explain how technological advances come about, raising the productivity of an economy. Those models say that educated researchers and investment in R&D are what generates technological improvements, which in turn boosts economic growth.

  Solow was unconvinced by some of the assumptions of endogenous growth, particularly in its simplest form, known as the AK model. (The ‘A’ in the title of the model refers to the economics shorthand for technology, while ‘K’ refers to capital.) This theory says that the rate of technical improvement in an economy is proportional to its growth rate; in other words, technology and the economy grow at the same rate. Solow thought that process seemed too neat to be plausible. Although they differ in terms of how growth comes about, these models follow the implications set out by the Solow model. Endogenous growth theories extend Solow’s neoclassical model in spelling out how innovators produce technological progress.

  Another criticism relates to the work of Douglass North discussed in the previous chapter. A difficulty of the Solow model is that it can account for differences in growth rates across countries only by appealing to technological progress. So, institutions such as those favoured by North play little role in explaining why some countries are wealthy while most are not.

  On the other hand, the Solow model can explain why countries have different levels of per capita income, and indicate whether we are converging to a slow-growth future. Growth should speed up if an economy is operating below its steady state, or the level of output that it is capable of producing. So, if an economy is starting to develop and has low levels of capital stock, then it should realize higher returns to its capital than a country which is developed and has had a lot of capital accumulation. If these economies have the same levels of technology, investment rate and population growth, then the developing country will grow faster than the developed one because of diminishing returns to capital discussed earlier. Th
e output per worker gap between these countries will narrow over time as both economies approach the steady state. This important prediction of the neoclassical model is known as the convergence hypothesis: developing countries will grow faster than developed countries if they have the same steady state until they converge to the same income level.

  Does this bear out empirically? If there is convergence, then there should be an inverse relationship between a nation’s starting level of income and subsequent growth. Japan, which started at a much lower level of development in the post-war period, grew more quickly than other more developed economies. From 1950 to 1990, Japan experienced growth that was on average much faster than that of the US. For rich countries there was an inverse relationship between their initial level of per capita income and growth rate between 1880 and 1973. However, there is no clear relationship in more recent periods for either rich countries or all countries in the world. So, there is limited evidence of convergence.

  Some poorer and middle-income countries (particularly China) have grown faster, and begun to catch up with wealthier nations, which is what the model predicts. But there are many poor countries that have grown slowly. In terms of the world income distribution, instead of seeing convergence, there have been signs of polarization between rich and poor nations.

  What about those nations that are developed and experiencing a slowdown in growth? What can be done to raise productivity in advanced economies? It is a question that other nations may also eventually face.

  The productivity challenge

  The Organisation for Economic Co-operation and Development (OECD) highlights the productivity challenge as one of the biggest issues since the 2008 banking crisis.8

  Britain is among the worst affected. By any number of metrics, UK productivity (output per hour) is lower than it should be based on pre-crisis trends, which is a puzzle. In other words, productivity growth has slowed down considerably since the crash.

  One way to think about the immediate post-crash period is that it has been a job-rich recession. Employment recovered a year earlier than output, and unemployment never hit the 3-million mark reached during the recessions of the early 1980s and 1990s. But output per worker was lower during this period since less output was demanded in a recession than in normal times. Since the 2008 crisis, output per worker grew at just 0.2 per cent per year, which is a fraction of the 2.1 per cent average growth rate between 1972 and 2007. Wage flexibility helped to maintain jobs during the latest recession, a decline in real wages making it possible to keep people in work.9

  Employers hoarding workers instead of laying them off doesn’t explain the entire productivity puzzle.10 Part of the answer may be that the British economy has a large services sector in which it is difficult to measure either investment or output accurately.11 But the US also has a large services sector and doesn’t suffer from the productivity problem to the same extent, so mismeasurement is unlikely to be the whole story either.

  The Bank of England has concluded that output per hour is around 16 per cent lower than expected.12 Unlike previous recessions, productivity hadn’t picked up during the recovery. This is the essence of the ‘productivity puzzle’.

  That said, productivity growth was already slowing down before the crisis. The OECD points to low investment as an explanation. As a share of GDP, UK investment began to trail that of the US, Canada, France and Switzerland in the 1990s. Investment fell from around a quarter of GDP in the late 1980s to just over 15 per cent. Low investment means there’s less productive capital for employees to work with, and thus lower output per worker.

  This was also one of the conclusions of the Bank of England. They can explain between half to three-quarters of the productivity puzzle. Mismeasurement accounts for around a quarter. They then looked at cyclical factors related to the business cycle and also at the possible structural reasons behind lagging productivity, i.e. how the economy is structured as opposed to cyclical variations. Some of the cyclical factors concern hoarding workers and doing work that doesn’t immediately add to output. Structural reasons include low capital investment and inefficient resource allocation, where workers are not moving from low- to high-productivity sectors. That can happen when there are high firm survival rates resulting in so-called zombie firms that have survived only due to the extraordinarily low interest rate environment.

  This is not just a UK problem, however. The term ‘secular stagnation’ has been revived as a concern for all developed economies and requires revisiting our models of growth. The slow recovery of the United States was what led Harvard economist Lawrence Summers to warn about a slow-growth future for advanced economies. At the forefront of this issue is Japan. Since its early 1990s crash it has experienced several ‘lost decades’ of growth, not helped by the survival of zombie firms in its initial recovery, which contributed to those unproductive years. Since then, Japan has launched the most aggressive economic policy in the world in an attempt to end decades of stagnation. As the rich economy with the most aged population, which is an important factor contributing to secular stagnation, how Japan fares will hold lessons for others.

  Japan’s ‘lost decades’

  Japan’s growth since the early 1990s has hovered around 0–1 per cent and productivity growth has been poor. The three major economic policy ‘arrows’ introduced by Japanese prime minister Shinzo Abe at the end of 2012 with the aim of revitalizing the world’s third largest economy have been dubbed ‘Abenomics’.

  The first arrow – aggressive expansion of the money supply in an attempt to end deflation or price declines – has failed to hit the target consistently. There have been positive signs, but the challenge of ending years of stagnant prices is immense. Stock markets have hit multi-year highs but the real economy hasn’t benefited sufficiently. Higher market valuations alone have not been enough for firms to raise wages that are fundamental to sustaining price rises. They are instead looking for more output per worker in order to justify higher pay. Average real wages were hit by the 2008 crisis and are yet to recover fully.

  The second arrow, fiscal policy, hasn’t quite hit the mark either. In one instance, a 2014 government decision to raise the sales tax from 5 per cent to 8 per cent, the first such increase in seventeen years, squeezed spending and tipped the economy back into recession. In its immediate aftermath, GDP contracted at an annualized pace of 7.3 per cent in the April–June quarter, the worst contraction since the economy shrank by 15 per cent in the 2008 global financial crisis. This mirrored 1997, when a sales tax increase had sent the economy into recession, revealing an underlying weakness of demand. Raising taxes was intended to reduce Japan’s indebtedness. It was always the case that any attempt to address the country’s staggering debt, which at around 240 per cent of GDP is the highest in the world, would be an economic drag, but the scale of it is a reminder of the fragility of the revival of the Japanese economy.

  The hardest arrow to score with was always going to be the third – the structural reforms that target the way that the economy is constituted and run. How can Japan raise productivity when its population and labour force are shrinking? Can firms be enticed to invest in a country where consumption demand is low after years of stagnation have taken their toll and people are concerned about taking on debt? Abe’s structural reforms include over 240 initiatives to raise productivity. Such reforms take time, and ministers warn that it could take a decade for Abenomics to work. So it may be years before the positive impact of any structural reforms are felt. Abe is Japan’s sixth prime minister in ten years. Time is seemingly a luxury for Japan’s leaders, yet it’s the very thing that they need to turn around an economy that’s been struggling for decades, during which Japan has fallen from the world’s second largest economy to its third.

  The country that overtook it also faces slower growth and an ageing population. For a middle-income country, China has a demographic profile that is similar to rich nations. Its working-age population is shrinking, though it has en
ded its ‘one child policy’ to counter the ageing demography. Also, if Britain and America as well as Japan are counting on innovation to keep them rich, China needs to get there before its growth slows down, as discussed in previous chapters.

  For Europe, the focus is also growth, and a lot is hanging on the governments’ ability to deliver. German Chancellor Angela Merkel has said that the legitimacy of the European project depends on people becoming better off. So the European Union is also focused on raising growth through investment, as discussed in the chapter on John Maynard Keynes.

  Raising growth and productivity to counteract a stagnant future is, then, a common challenge for major economies. And it is one that has been increasingly recognized by policymakers. Turning to Great Britain, the government has begun to focus on economic growth, particularly with respect to its particularly worrying low-productivity challenge.

  The UK government’s renewed focus on growth

  During the 2008 banking crash, ‘benign neglect’ of the productivity issue by successive governments who were focused on the immediate crisis meant there was insufficient attention paid to economic growth. As the country worst affected by the global productivity slowdown, Britain has since placed this issue at the centre of its economic growth agenda. For one thing, following research by the Bank of England and others, the government has focused on raising investment.

  For instance, it has established a National Infrastructure Commission. The UK needs investments in ‘hard’ infrastructure (such as transport links) as well as ‘soft’ (such as digital networks), which can be just as important to induce business investment. Britain’s track record on this issue has been somewhat mixed. Development of the digital economy has been impressive in some respects. For instance, Silicon Roundabout in London has attracted more venture capital than other European cities. But there are also areas of the country where even getting a mobile phone signal is challenging. The other significant area where investment is needed is skills. Business surveys routinely point to a skills shortage cramping their growth. It seems that investment is needed in physical and also digital infrastructure as well as in human capital.