(This essay was published in Hong Kong Economic Journal on 10 May 2017.)
The emergence of Donald Trump and Bernie Sanders from outside the American political mainstream in last year’s presidential election reflects the rise of right wing and left wing populist reaction to poor economic performance – both low productivity growth and high economic inequality. Economic factors may not be the only reason for the rise of populism in the US and in the rest of the world, but it is certainly a major part of the explanation.
It is not without irony that the president of communist China is now championing a liberal world economic order and is telling the president of capitalist America the benefits of free trade. The irony is perhaps not surprising since China sees itself as still rising economically and a beneficiary of a liberal economic order, while America worries about economic decline and being a victim of the open global economic order it once championed.
Ever since the 2008 financial market meltdown, it has been customary to blame global economic integration and free market capitalism for failed economic performance. This accusation has gained widespread popularity in the media and public policy debates. But is it justified?
I do not think so and here is why: the wave of global economic integration and free market capitalism did not really start until after 1980, and mostly towards the latter part of the 1980s. But economic performance in the West, particularly in the US, started to weaken in the 1970s. The forces behind this slow weakening have remained in place up to the present time.
From ‘Left of Center’ to ‘Right of Center’
The first time global free market capitalism was faulted for failed economic performance was in the first-half of the 20th century. It was widely believed that the global economic integration in the latter half of the 19th century imploded on itself during the Great Depression of the 1930s and the two world wars. The resulting economic devastation greatly damaged confidence in capitalism. Capitalism was also severely challenged by the triumph of communism in Russia and the rapid industrialization of the Soviet Union.
In the aftermath of all these events, the dominant intellectual ideas leaned more towards socialism than capitalism. Most intellectuals in the West held ‘left of center’ views. In the economic domain, they embraced pro-government intervention ideas associated with Keynesian economics. In politics, social liberalism, progressivism, and social democracy dominated the landscape of the West.
By the 1960s, the ‘left of center’ view came to dominate academia in North America and Western Europe and became the ‘centrist’ view by default. It was identified as the new face of liberalism in the 20th century. The few academics with views that were ‘right of center’ found the intellectual climate in the universities quite hostile, and felt marginalized.
This intellectual climate began to change only after the two oil crises of 1973 and 1979. Mainstream Keynesian economic ideas at the time failed to provide a credible explanation for the new phenomenon of stagflation. Milton Friedman and the Chicago School provided a better alternative explanation that convinced many in the economics profession. This laid the foundation for public policy opinion to shift to ‘right of center’ pro-market policies.
Pro-market policies only began to gain ascendancy after Margaret Thatcher became UK Prime Minister in 1979 and Ronald Reagan became US President in 1981. Of even greater significance on a global historical scale was Deng Xiaoping’s decision to reform and open China’s state-dominated economic system in late 1979. A new wave of global economic integration and free market capitalism began in the 1980s.
Technology Impacts Economic Performance
Professor Andrei Shleifer of Harvard has aptly called the period 1980-2005 the ‘Age of Milton Friedman’. This means the impact of ‘right of center’ policy ideas could hardly have been felt in the US until the late 1980s. The impact of technological changes on economic performance, however, had already been felt as early as the 1970s.
Research conducted by Professor Kevin Murphy showed that the real wage rates of men grew by 19.4 per cent in the 1940s, 29.7 per cent in the 1950s, 24.1 per cent in the 1960s, and only 5.0 per cent in the 1970s and -7.8 per cent in the 1980s (see Figure1). In other words, labor productivity growth had started to decline rapidly in the 1970s and turned negative in the 1980s.
Robert Gordon’s work The Rise and Fall of American Growth (2016) reconfirms these results and shows that US labor productivity was at its peak during 1920-70, but fell off significantly during 1970-2014.
The slow growth of labor productivity of 1.62 per cent per annum during 1970-2014 compares unfavorably with the 2.82 per cent during 1920-1970. Gordon shows that the change in labor productivity is due mainly to changes in total factor productivity, which represents innovation and technological change.
The reason for this slowdown is that most technological advances since the 1970s have tended to be channeled into a narrow sphere of human activity involving entertainment, communication, and the collection and processing of information. This narrow focus explains why the new technological advances have not had a big economy-wide impact in lifting productivity. They have benefitted only a small number of industries in the economy and a limited, highly-skilled fraction of the workforce.
The economic slowdown in the US preceded the shift to ‘right of center’ pro-market economic policies by more than a decade. Global economic integration and free market capitalism in all likelihood prevented productivity growth from declining faster. The earnings growth of companies like Apple would have been less spectacular if their products had not been largely manufactured and assembled in China.
If the technological advances since the 1970s have been concentrated in only a small number of sectors, then could this have worsened inequality? In the period 1979-90 when global economic integration had hardly taken off, the weekly earnings of US college graduates increased by 2 per cent while the real weekly earnings of high school graduates decreased by more than 16 per cent.
As a result, the wage premium for college graduates over high school graduates increased from 42 percentage points in 1979 to 71 percentage points in 1990. Consequently, overall wage inequality for men grew dramatically between 1979 and 1990. Wages for men at the top end of the wage distribution grew by 18 per cent relative to wages for men at the bottom of the wage distribution over the 1980s.
The pattern of economic inequality in the US also changed over time from 1939 to 1989. Wage inequality among men rose in the 1970s and 1980s, but it fell rapidly in the 1940s and was very stable in the 1950s and 1960s. What explains these changes?
Again, research conducted by Professor Murphy showed that technological advances in the 1980s and 1990s were biased towards skilled workers, in the 1950s and 1960s the advances had neutral impacts between skilled and unskilled workers, and in the 1940s they were biased towards unskilled workers.
This means that during the 1970s and 1980s, technological advances demanded more skilled workers and labor markets were paying a premium for skilled services, thereby increasing wage inequality. In the 1940s technological advances demanded more unskilled workers and labor markets were paying a premium for unskilled workers and thereby decreasing wage inequality. In the 1950s and 1960s, technological advances did not favor either skilled or unskilled workers so wage inequality remained stable.
The changing pattern of technological advances over the five decades provides a more convincing explanation for the patterns of wage inequality and productivity growth over this period than blaming it on global economic integration. Failure to appreciate the true forces behind the decline of productivity growth and rising wage inequality will lead to the adoption of wrong policies that will not solve the problems society faces. They may even worsen them.
It is unlikely that we can control technological advances to become biased towards either skilled or unskilled workers. So the best policy is to influence the supply of skilled versus unskilled workers to support productivity growth and mitigate the effects of wage inequality. Wrong policies will not get us to the right solutions.
Ideological debates will always continue, but it is important that we bring scientific analysis and empirical evidence to illuminate the questions we wish to resolve. Shouting matches driven by deep convictions add little value. In the age of social media, where communications channels are echo chambers for the convinced and converted, shouting matches add no value and only reinforce prejudice.