top of page
  • Writer's pictureeconomictribune

Live to work or work to live? Examining the returns to longevity in Singapore

By Shaun Ng, Research Fellow

Growing up in Singapore, I remember the compulsory social studies modules depicting Singapore's history and the narrative of its transformation from a “Sleepy Fishing Village” to a First World economy. As a city-state deprived of any natural endowments besides its human capital, labour productivity has been and is a key driver of growth and competitiveness in the global economy, echoing Paul Krugman’s aphorism: “Productivity isn’t everything, but in the long run, it’s almost everything.”

Singapore has often been touted as a country with one of the highest life expectancies by numerous international organisations such as the World Bank and the United Nations. Singapore’s life expectancy, measured at birth as the average number of years a newborn is expected to live at current death rates, has increased steadily from 65 years in 1960 to 84 in 2019. Correspondingly, in the same period, Singapore’s labour productivity has soared from a measly $3.90 to $54.55 per hour worked. While many studies have been done to decompose the drivers of labour productivity growth, there are fewer studies devoted to uncovering the interdependence between these two issues. Could Singapore’s major labour productivity improvement be in part a result of its rising life expectancy?.

This question remains theoretically ambiguous: on one hand there are many channels through which life expectancy may raise productivity, such as increased educational attainment and higher precautionary savings levels raising investment outlay. On the other hand, there might be an onset of Malthusian effects whereby life expectancy may raise population growth, suppressing labour productivity.

In economic research, life expectancy is commonly used as a measure of health outcomes. And in policy, life expectancy, which is commonly associated with ageing, is often angled as an assessment of demographic risk—i.e. risks associated with changing population age, gender, socioeconomic status, to inform policy design.

In economic literature, there is much debate about the direction of the relationship between life expectancy and labour productivity, with most acknowledging that both reverse causality and simultaneity are likely to hold. For instance, more productive workers might command higher wages which increases the financial accessibility of greater education attainment, which imply a backward effect of the education channel outlined earlier. Furthermore, it is conceivable that these two issues are contemporaneously determined, perhaps through underlying health systems and pension schemes. In my attempt to mitigate endogeneity, I adopted a 2SLS model with lagged life expectancy as the instrument for present life expectancy, arguing that present labour productivity today is unlikely to have affected life expectancy in the past. Furthermore, I conjecture that life expectancy in the past is likely to correlate with present life expectancy through the existing health systems and demographic factors such as gender balance and educational systems.

While there is less ambiguity in adopting expected life expectancy at birth as the metric for life expectancy, the selection of measurement for labour productivity is more complicated. How should we measure output? Should we measure output averaged across the entire population or only the labour force? How do we account for non-full-time workers? Different measurements provide very different insights attributable to the unique interactions of life expectancy on worker demographic. Ultimately, I decided on output per hour worked to limit the scope of analysis to that of the workforce. Firstly, this excludes the effects of life expectancy on the share of the retired population and its drag on aggregate productivity measurements. Secondly, this removes variation in productivity stemming from the number of hours worked per worker which is likely to be widely correlated with the fluctuations of the business cycle.

During my literature scan, what was interesting was the wide range of findings on the relationship between life expectancy and growth or productivity - while most authors found a positive relationship, some argued for a negative relationship. In contrast, others suggested that they have no relationship. These differing results are largely attributable to the differences in methodology employed, sample selection of countries, measurement units of growth or productivity, and time period of study which incorporated epidemic shocks of varying severity and persistence.

A study that particularly piqued my interest however was that by Cervellati and Sunde (2011), where their findings suggested that a U-curve relationship existed, which was dependent on whether a country has experienced a “Demographic Transition”—i.e. when it experiences a declining fertility and birth rate, coupled with an expected life expectancy at birth exceeding 50 years.

From a cursory review, my research findings suggest that life expectancy does indeed have a positive effect on labour productivity. A percentage point increase in life expectancy is associated with an approximately 3.7% increase in output per hour worked on average, and is considerably higher than the other controls in the study such as young and old age dependency ratios. This result is likewise in line with that of Cervellati and Sunde (2011) whereby once such a transition has occurred, the country is unlikely to experience the onset of Malthusian effects.

I caveat that the results should be interpreted with caution. While this research has certainly failed to control for all numerous factors that affect labour productivity, it did however shed light on the relationship between health outcomes and their contribution to strengthening productivity in Singapore. With a rapidly ageing population and a sharp decline in the resident labour force coupled with historically low fertility rates, perhaps we can consider labour productivity gains as a potential policy outcome of improvements in healthcare systems in addition to a demographic risk management solution. It would also be worthwhile to analyse both labour productivity and life expectancy as separate outcomes to identify drivers that may potentially raise both life expectancy and labour productivity.

Thomas Edison once said, “There is no substitute for hard work”. Maybe there is—just live longer.



Cervellati, M. and Sunde, U. (2011). Life expectancy and economic growth: the role of the demographic transition. Journal of Economic Growth, 16(2), pp.99–133. doi:

World Bank (2022). World Development Indicators | DataBank. [online] Worldbank. Available at:

Feenstra, R.C., Inklaar, R. and Timmer, M.P. (2015). The Next Generation of the Penn World Table. American Economic Review, 105(10), pp.3150–3182. doi:


bottom of page