Published Date: 27 August 2020

“The Global Jobs Crisis and Why We Should Think Longer Term” – Remarks by Mr Tharman Shanmugaratnam, Chairman, Monetary Authority of Singapore, at the Jackson Hole Economic Policy Symposium hosted by the Federal Reserve Bank of Kansas City on 27 August 2020

I.     Confronting concurrent realities

We are on uncharted terrain. COVID-19 is a historic challenge, but we are really faced with three testing realities at the same time.

We have the immediate challenge of a public health and economic crisis, both of global scale. But tackling this is complicated by the second reality - that of a new future that is already unfolding before us, for which we have to re-gear ourselves. The shape of demand, supply chains and the way work is organized are all changing, and with them the job market. COVID-19 and the fear of future pandemics have brought on a new normal that we must adapt to.

Third, we inherit long-standing legacies, now made more challenging by the COVID-19 crisis: the stagnation of middle incomes in the advanced world, widening inequalities, the growing prospect of pension gaps in many countries, and an ever more serious climate crisis.

It is these three realities interacting together that make this an immensely complex, dynamic and daunting time — for public policy, for society, and for politics everywhere. Focusing on today’s problems in themselves, without both addressing the problems we inherit and equipping ourselves for a structurally different future, will make the years to come even more testing, and fraught with danger.

I.A. Too early for creative destruction

Policy-makers have passed the first phase of tackling the economic crisis, where most nations were focused on avoiding the extreme downsides arising from COVID-19 and lockdowns. We wanted to avoid mass unemployment and a large wave of bankruptcies, which would have led to a self-reinforcing spiral of declining incomes and public spirit. We have mostly succeeded in doing so.

We are now at a transition point. We cannot protect existing jobs, firms and industry capacity to the extent seen in the first phase. Most countries do not have the fiscal capacity to do so. Strategically too, we have to let some of today’s firms and jobs go, to enable new growth and more promising jobs to be created. We have to allow for ‘creative destruction’, to secure vibrant economies over the longer-term.

However, we cannot pivot policies too quickly into this next phase. Our economies are still in significant slack, the virus is far from tamed, and there is risk of repeated lockdowns especially during the northern winter. Even on optimistic assumptions, it will take much more than a year to get back to full economic life, and back onto a normal path of output and employment.

There is another reason why substantial macroeconomic policy support for firms and jobs remains necessary over the medium term. Withdrawing support too quickly will not lead to efficient market restructuring — the rechanneling of labor, capital and other resources that takes us to a better place — in the way we would expect in a typical downturn. We are operating with little visibility of the future, hampering the ability of either policymakers or markets to discriminate between firms that are loss-making today but remain viable for the future and the zombies that limp along on life support. Creative destruction without this visibility will lead to too many good firms and jobs being destroyed. 

The next phase of policy response must also consider the pre-existing trends and inequalities that COVID-19 has exposed and in fact accentuated. We must also make a determined attempt to grow out of the crisis and create jobs in a way that strengthens, and puts at the core, the sustainable economy.

II. The Central Problem: Jobs

The central economic problem of the times lies in labour markets.

First, there is a real prospect of prolonged high unemployment in many economies. It is not at all assured that we will get back to tight labor markets, even with traditional macroeconomic policies working the way they should.

Second, we have to counter the secular decline in productivity growth, to be able to tackle the stagnation of median incomes seen in a whole range of advanced economies, with a few like Sweden and Singapore being the exceptions. It cannot be achieved without reviving productivity growth.

Third, we have to address the trend towards a polarized job market — with more jobs being created at the high and low skilled ends of the labor market and fewer in the middle, and the growth of an insecure gig workforce, especially amongst the young (Haldane 2019; Boeri et al. 2020).

Each of these is a difficult challenge, and overcoming all three even more so. While employment recovered well in some major economies in the decade after the Global Financial Crisis, productivity growth weakened (and indeed was part of the reason why employment recovered) and job polarization became more severe. 

There have been some periods in history where we were able to achieve all three objectives. One of the stylized facts of economic development, which Nobel Laureate Arthur Lewis and other economists highlighted, consisted of surplus labor being freed up from low-productivity sectors (originally agriculture) and taken up by sectors which had higher productivity such as manufacturing. That reallocation of labor is how low-income countries became middle income, and some middle-income countries became advanced. 

But the experience of recent years in many advanced economies has been of a reverse Lewisian movement — where labor is shed from industries that are dynamic and are at the higher end of the productivity ladder, and channeled into lower-productivity, lower-wage occupations. In fact, the main engine of job growth in the United States and several other countries has been in lower-paid services — in F&B, cleaning, security, and a range of other domestic services.

COVID-19 has introduced a major disruption to this already polarized, barbell-shaped market. Remote work, telepresence — like how these Jackson Hole discussions are taking place — are shrinking the number of jobs in a whole range of domestic and personal services.

What this means is that the market receptacle for labor displaced from higher productivity industries has shrunk. It was a receptacle of lower-pay and less secure jobs, but what is even more unattractive now is the prospect of fewer such jobs. 

We also face, in particular, the unravelling of some of the gains in women’s participation in the labor force and sense of self-determination that were achieved over decades. COVID-19 and the new protocols that we will have to live with for some years is hurting the human-facing jobs in the services sector, that women have been disproportionately represented in — brick-and mortar retail, food services, hotels, and the leisure industry. Compounding this has been the closure of childcare and schools during the pandemic, which has unfortunately translated into more women than men staying at home to look after their kids.

We have to apply ourselves now and over the next few years to this central problem of jobs: making sure we have enough jobs, and that people do not get trapped over time in low skill, low pay work.

II.A.  Widening the distribution of good jobs

What are some of the initiatives that we have to focus on, or begin thinking about? I will touch on macroeconomic policy at the end of my remarks, but we have to focus not only on macro-economic stimulus, but increasingly on the microeconomic incentives we create, and on new forms of collaboration between government, business and communities. I highlight a few below, that are essential to tackling the jobs challenge.

First, we have to redesign fiscal support for firms and jobs, so as to provide strong incentive for new job creation as we gradually phase out subsidies for existing jobs. In other words, we should shift from subsidizing the stock of existing jobs towards incentivizing the flow of new jobs. Singapore is doing so with our new Jobs Growth Initiative (the state pays employers up to 50% of wages for new additions to the firm’s workforce over the first 12 months), as we step down the significant subsidies for existing jobs that our Jobs Support Scheme provided in the first phase of the crisis.

Second, we have to put great energy into the coordination needed to help people who are displaced to get back into jobs, and to avoid as much as possible ill-fitting job matches that could lead to a permanent reduction in their wages, as well as in average productivity.

We must do all we can to avoid people being detached from work for an extended period, to prevent the depreciation of skills and morale that comes with that, and increased barriers to their employment. Economists call it ‘hysteresis’. It is a real risk in the coming months and years.

However, the aim of job matching cannot simply be about speed. It is not about getting someone any job as soon as possible, because that can mean losing the value of the skills they have accumulated. As much as possible, we have to find jobs that build on a person's skills and experiences. A key coordination task, therefore, is to find those jobs with adjacent or complementary requirements, and to move swiftly to top up the skills that the job-seeker brings. We cannot take too long to help someone secure a new job, but we have to strive for a good match with the human capital they have accumulated, so that they do not lose it, and society does not lose it.

Getting to this outcome is not easy, because labor markets suffer from imperfect information. It requires joining up career coaching, job matching and skills development programs. This is a very human enterprise, but greatly aided now by new ICT and AI tools. The most effective systems have involved official agencies playing an active coordinating role, working with employers, job-seekers, unions or other social partners, and training providers. Even neighborhood-based referral networks have been found to be useful in securing reemployment, especially for lower-paid workers (Hellerstein et al. 2019). However, in the context of the current uncertainties, significant state subsidies are also required to encourage employers to recruit and train people. If we leave it to the market, we will see longer spells of joblessness.

We have to recognize too, in the wake of the pandemic, that it will take time before hiring for permanent jobs comes back on the scale we need. We have to incentivize firms to take in people on attachments of various forms, even if they cannot take on permanent hires. This means not just youth apprenticeships, but traineeships for mature workers in their 40s and 50s. We have to work with industry partners to curate traineeships, get people back into the workplace, and help them to add to their skills. We must find every way to avoid a long-term detachment from work, and the permanent scarring that comes with that. 

Third, we have to invest more systematically in both education and life-long learning to tackle the more fundamental, pre-existing trend of job polarization. We have to go beyond the mantra of life-long learning to make it a practical reality for blue-collar and ordinary white-collar workers. All experience, even in northern Europe where the tradition is most established, shows that it has been much harder to get this going for ordinary workers than for high-skilled professionals, which only accentuates existing inequalities.

That has to be a core focus of public-private collaboration – developing quality training options that workers find appealing and relevant to their careers, and using technologies and community outreach programs that make learning convenient when they are not on the job. It also requires increased agility in the system, to enable firms and workers to develop new skills quickly as the economy restructures. 

Fourth, again to create enough good jobs and avoid continued job polarization, we have to ask ourselves questions about incentives for different kinds of automation. Tax policies today incentivize automation in almost every form and everywhere, even where labor markets face significant long-term slack. Should we incentivize automation that augments labor, creating new tasks in which humans have a comparative advantage, rather than simply displacing labor (Acemoglu and Restrepo 2018)? 

Technological advances have historically both displaced workers and increased the demand for labor in new areas — either within the same industry or indirectly in the rest of the economy. It has enabled society to stay fully employed while productivity moved up. But that is an empirical regularity, not a law. Serious economists, as well as business and technology leaders themselves, have differing views on whether we can be equally sanguine about an emerging generation of increasingly potent technologies, or whether we are heading for a permanent state of higher joblessness. 

We will not know in advance. But we know that some new technologies, like collaborative robots, augmented reality and certain AI tools, generate more demand for human skills, including mid-level skills, than other advances. It is not too early to consider how public policy can encourage automation and skills development of the kinds that can complement one another, so we increase the chances of a new era of widely-distributed good jobs. 

Fifth, we need collectively-determined solutions to ensure that growth of worker compensation does not trend below productivity growth, as has happened over the last few decades in many economies. Studies show that a range of market imperfections could account for such outcomes. Institutional interventions, involving the state, are especially needed where the problem of low or lagging wages reflects either the weakened bargaining power of workers (Stansbury and Summers 2020), or increased monopsony power of employers. Ideally too, these interventions should be accompanied by efforts to develop new social and employer mindsets, so that new norms can persist in the labor market without requiring perpetual state intervention. Norms do appear to vary across countries, especially with regard to pay for women and less-skilled workers. There could be a role for monetary policy as well, as the late Alan Krueger suggested at Jackson Hole in 2018. If monetary policy can keep the labor market hot, it could help weaken the monopsony power of employers (Krueger 2018).

Sixth, we need strategies to ensure that the SME sector survives and adapts to the new, post-pandemic normal, to avoid much larger job losses in time as well as greater wage inequalities. SMEs are disproportionately represented in the service sectors, where jobs have been most vulnerable to the impact of COVID-19 and national lockdowns (Bartik et al. 2020). They also face greater funding risk than large firms in an environment of uncertainty. While a churn of businesses is inevitable and will have its economic benefits, there is a real risk of a diminished SME sector as a whole and further increases in industry concentration in many economies. Besides its implications for long term generative capacity and economic vibrance, it will likely leave an impact on social capital. 

Governments and central banks have been quick to move on loan guarantees and the like, to ensure near-term credit support for SMEs. But we have to look beyond this. There is no silver bullet here, but we do need to focus on broad-based economic vibrance as an objective of public policy, and be willing to experiment. Quite apart from updated competition policies, what form of cluster strategies can we get to work well, and to enable better diffusion of innovation from the frontier firms to the rest? How do we incentivize test-bedding of innovations in small firms themselves? How do we aggregate and organise skills training for their workers, to address the growing gaps in skill levels between large and small firms? How do we promote digital platforms — and especially inter-operable cross-border platforms — that small firms can plug into, so they can benefit from the scale economies of the digital economy?

III. Monetary and Fiscal Policies: Taking the Long View

Governments have done right to pull out all the stops to save their economies and jobs amidst the pandemic. However, we face a different structural future, compounded by the long-standing social and economic challenges that we inherit, that make this much more than a sharp cyclical downturn. It requires a longer-term orientation in economic policymaking and in the way we organize ourselves to respond collectively. This is a major and complex theme, but let me make a few broad remarks here.

Monetary policy has succeeded in buffering economies and financial systems against a severe shock to confidence amidst COVID-19 and a seize-up in liquidity, including in the global funding markets. But its very expansiveness has brought us closer to the point of diminishing returns to monetary policy as a demand management tool in the advanced economies. Relying chiefly on monetary policy to get economies back to full health will be less effective over time and holding policy rates at or near zero over a prolonged period may have counter-productive effects.

Fiscal policy therefore has to play a more important role, and in concert with labor market and industrial policy reforms such as those I mentioned earlier. But it also requires a repurposing of fiscal policy. We have to repurpose the state - to defy secular stagnation, to catalyze the green economy, and to rekindle social mobility. The state has to go about achieving these objectives with a spirit of activism, guided by moral purpose, and always focused on achieving both equity and efficiency.

Expansive fiscal policies will make sense for a while, probably a few years in most economies. But the real debate has to move from questions around immediate stimulus effects, to how we achieve our longer-term goals. We need a different kind of Keynesianism: an active state aimed at long-term recovery and regeneration, nationally and globally.

We have to refocus on public goods, which used to be a fundamental purpose of government budgets. They are now greatly underinvested in, not only in the emerging world but in most advanced economies. The pandemic was a harsh reminder of this, with public health systems overwhelmed in many countries. But this is now a huge opportunity around the world: investments to achieve broad-based quality in public education systems, including regularly refreshed technical and applied training; in more accessible and affordable healthcare; in universal broadband access; in upgraded water and transport networks; in R&D for clean energies, sustainable agriculture and the circular economy. 

We will not be able to fund these investments on public balance sheets alone. The redesign of fiscal strategy must also involve catalyzing private investment in public goods. There are many examples of how this can be done, including ways to achieve a fair sharing of the risks on such investments.

Remember that this pandemic is not the last we will face, and the next pandemic may very possibly be more challenging than what we are going through today. So we do need, once we have emerged from this crisis, to rebuild fiscal space to deal with future shocks – and at least avoid a continuing build-up of debt-to-GDP ratios. Most governments will have to strengthen tax revenues. They will have to do so while giving a fair deal to those with lower and middle incomes, and ensuring that fiscal systems meet the challenge of politically durability.

The repurposing of the state requires boldness of imagination, and the marshalling of public support around nations’ long-term goals. It will help rebuild broad-based prosperity and cohesive societies, and can put us on course to an ecologically viable future. It may be our only chance.


* * * *


Author’s Note: I am grateful for discussions with Edward Robinson, Andrew Colquhoun, Yin Sze Liew and Dingxuan Ng at the Monetary Authority of Singapore, and for their background research.


Acemoglu, D. and Restrepo, P. 2018. “The Race Between Man and Machine: Implications of Technology for Growth, Factor Shares, and Employment.” American Economic Review. Vol. 108(6). pp. 1488–1542.

Autor, D. 2019. “Work of the Past, Work of the Future”, AEA Papers and Proceedings, American Economic Association, Vol. 109, pp.1–32, May.

Bartik AW, Bertrand M, Lin F, Rothstein J, and Unrauth M. 2020. “Measuring the Labor Market at the Onset of the COVID-19 Crisis,” NBER Working Paper No. 27613, July.

Boeri T, Giupponi G, Krueger AB, and Machin S. 2020. “Solo Self-Employment and Alternative Work Arrangements: A Cross-Country Perspective on the Changing Composition of Jobs”, Journal of Economic Perspectives. Vol. 34 (1). pp. 170–195.

Haldane, A. 2019. “Climbing the Jobs Ladder”, Speech Delivered at Glanford Park Stadium, Scunthorpe, 23 July.

Hellerstein JK, Kutzbach MJ, and Neumark D. 2019. “Labor market networks and recovery from mass layoffs: Evidence from the Great Recession period,” Journal of Urban Economics, September.

Krueger, AB. 2018. “Reflections on Dwindling Worker Bargaining Power and Monetary Policy”, Speech delivered as the Luncheon Address, Jackson Hole Symposium.

Stansbury, A and Summers, L. 2020. “Declining Worker Power and American Economic Performance,” Brookings Papers on Economic Activity, March.