Whither Singapore’s Unemployment Rate? by Hoon Hian Teck, Singapore Management University In the period 1990-99, Singapore’s average annual real GDP growth was 7.3 per cent with a coefficient of variation, which is a measure of dispersion of growth around its mean, of 0.49. 1 The average annual total unemployment rate during this period was 1.9 per cent. 2 In comparison, in the period 2000-12, when the average annual real GDP growth was 5.6 per cent with a coefficient of variation of 0.79, the average annual total unemployment rate was 2.8 per cent. Looking into Singapore’s next 50 years, if real GDP grows at 3 per cent on average, what does it imply for the rate of unemployment? Will slower growth coincide with more variable growth? This essay explores these questions? Before proceeding to explore what slower growth might mean for the unemployment rate, let’s review why Singapore is expected to grow more slowly. It helps to begin with a simple accounting relationship: Growth rate of total real GDP is the sum of the growth rate of real GDP per worker and the growth rate of labour force. Real GDP per worker is a measure of average labour productivity so we can say that total GDP growth rate is the sum of the growth of labour productivity and labour force growth. The first three decades or so of Singapore’s growth is best described as catch-up growth when labour productivity grew by racing to catch up to the world technology frontier, a process facilitated by its business friendly environment which attracted multinational corporations to base their manufacturing activities here to produce and sell into the world market. This process of convergence involved moving up the value chain in manufacturing. Available data show that, in the past decade, the employment share of manufacturing has shrunk while the share in the services sector has expanded. While the current restructuring effort is aimed at boosting productivity across all enterprises in the services sector, the experience in the developed economies shows that the pace of productivity improvement is more muted in the services sector. A frequent reference is made to the fact that the rate of growth of real GDP per person in the United States (U.S.), a world technology leader, has averaged 2 per cent over the last century. Even so, many economists now believe that future labour productivity growth in the U.S. will be

1

By real GDP growth, we refer to the value of gross domestic product measured at constant market prices. The calculation here is based on the time series on GDP at 2005 market prices available from Singapore Department of Statistics. Formally, the coefficient of variation is calculated as the standard deviation divided by the mean. 2 The data on unemployment is available from Singapore Ministry of Manpower.

1

less than 2 per cent. 3 If Singapore can manage to achieve labour productivity growth of 2 per cent per annum and the labour force grows at 1 per cent on average over the next 50 years, total real GDP growth will average 3 per cent per annum. Will the total unemployment rate rise above 3 per cent? The framework that economists use to think about the determinants of the unemployment rate is the search-and-matching model of the labour market. 4 At the heart of the model is the decision that firms make to create job vacancies and to recruit suitable workers to fill these vacancies. If firms are free to create job vacancies so long as they are willing to incur the necessary costs to recruit workers, then the equilibrium condition that determines the tightness of the labour market equates the expected cost of recruitment to the surplus that the firm expects to gain from employing another worker. 5 The surplus, in turn, is given by the excess of the present value of the worker’s marginal labour productivity over the present value of the worker’s wage. Using this equilibrium condition, we can draw a few inferences about Singapore’s future unemployment rate. First, the adage that “wage increases need to be matched by productivity increases” holds in the model. The result of having wage increases that outrun productivity gains is that there is a rise in the rate of unemployment. The experience of Western Europe provides a cautionary tale. France, Germany and Italy all had low unemployment rates in the 1960s – about 2 to 3 per cent. However, in the succeeding decades, the average unemployment rate in these economies steadily ratcheted up with France and Italy experiencing double-digit unemployment rates by the start of the new millennium. One explanation points to how the rapid growth due to the rebuilding of these economies after the Second World War was not recognised to be transitional so that when the process of convergence ended the economies would necessarily slow down. As a result, the high growth expectations led to wage increases that were ultimately not met by actual productivity gains. As a consequence, the unemployment rate began to rise in Western Europe. 6 One lesson for Singapore is that the

3

See John G. Fernald and Charles I. Jones, 2014, “The Future of U.S. Economic Growth,” American Economic Review: Papers and Proceedings, Vol. 104 (5), pp. 44-49 and the references cited therein. 4 The Nobel prize for economics in 2010 was awarded to three economists who developed the search-andmatching model: Peter Diamond, Dale Mortensen, and Christopher Pissarides. 5 Formally, the tightness of the labour market is given by the number of job vacancies for every unemployed worker. The tighter the labour market is, the lower is the rate of unemployment. 6 One might ask why the unemployment rate remained stubbornly high in Western Europe since, with the passage of time, workers would come to recognise the growth slowdown. An explanation is that institutions that provided strong support to the unemployed interacted with the growth slowdown to keep the

2

workforce must adjust to an era of slower growth. On the other hand, if efforts continue apace to generate innovation and achieve 2 per cent of labour productivity growth, workers can look forward to real wage increases of 2 per cent without rising unemployment. Second, successful matching of workers with the right skills to meet the needs of the new jobs in the next half century will enable the unemployment rate to stay low. The modern economy is fraught with fresh novelties thrown up by new technologies as well as ideas of business people. 7 In a sense, this means that the type of skills needed for jobs of the future cannot be accurately forecasted. Yet, our educational institutions can seek to teach people how to learn so that they thrive in an environment fraught with novelties. Additionally, while it is the private business enterprises that will create most of the new jobs, the government can facilitate a close communication between businesses and training institutions so that supply can match the demand for new skills. Third, it appears, judging from the experiences of today’s developed economies, that the era of slower growth will be accompanied by increased volatility. Without strong growth to provide a buffer, negative external shocks could translate into more lost jobs. Recessions could become deeper. What tools should be used to fight recessions? Since 1981, the Monetary Authority of Singapore has adopted an exchange rate-based policy rule, adjusting the exchange rate according to how far the inflation rate deviates from an implicit inflation target and the output gap. 8 This has led to reduced volatility in inflation and output. 9 However, during major recessionary episodes such as the 1997-98 shock from the Asian financial crisis and the 2008-09 Lehman Brothers crisis, major cuts in wage costs were implemented to fight the recessions. In order to finance jobs credits during episodes of negative external shocks, the government would need to save during good times. Even though the phase of catch-up growth is over, mature economies sometimes face prolonged periods when economic activity picks up – such as the U.S. internet boom in the second half of 1990s – and tax revenues increase (at given tax rates). A fiscally prudent government will

unemployment rate high. See Olivier J. Blanchard, 2006, “European Unemployment: Evolution of Facts and Ideas,” Economic Policy, Vol. 21 (45), pp. 5-59. 7 See Edmund S. Phelps, 2013, Mass Flourishing: How Grassroots Innovation Created Jobs, Challenge, and Change. Princeton: Princeton University Press for an emphasis on the ideas generated by people engaged in the world of business. 8 The output gap refers to the deviation of actual GDP from potential GDP. 9 See Ilian Mihov, 2013, “The Exchange Rate as an Instrument of Monetary Policy,” Special Feature A, Macroeconomic Review, April 2013, Monetary Authority of Singapore, pp. 74-81.

3

save up these additional fiscal resources in order to use them to hasten economic recoveries when negative shocks hit the economy. Singapore’s first 50 years saw the economy deliver economic growth that very likely exceeded the expectations of its workforce. As a result, productivity growth exceeded wage expectations thus leading to a steady decline in the unemployment rate. In the next 50 years, Singapore can avoid the way that the unemployment rate ratcheted upwards in Western Europe since the 1960s. To do so, it must be innovative to deliver the needed 2 per cent labour productivity growth so that real wages can increase at 2 per cent per annum. Its ability to facilitate the matching of workers to jobs will also help to keep the unemployment rate low. Finally, running budgetary surpluses in good times will enable the government to have the fiscal resources to fund jobs credits to save jobs in bad times.

4