Wednesday, May 31, 2023
Question 1: What will be the implications of recent developments in AI on global economic growth, as they mature over the upcoming decade?
Question 2: What will be the implications of recent developments in AI on unemployment in high income countries over the upcoming decade?
Contact us for more information
| Participant | Answer | Confidence level | Comment |
|---|---|---|---|
Ethan Ilzetzki
London School of Economics |
Remain unchanged | Not confident at all | |
Łukasz Rachel
UCL |
Increase (to 4-6%) | Not confident at all |
It is extremely hard to be confident about these issues. The range of estimates I have seen, e.g. from Goldman Sachs and co, is very wide. And the impact through political process, GE effects of more uncertainty, etc, is very hard to predict. |
Jagjit Chadha
National Institute of Economic and Social Research |
Remain unchanged | Not confident at all |
it is another innovatation on the long road of economic progress. it will throw open opportunities and problems in equal measure. The ultimate outcome in terms of growth and its distribution will depend on which policies are adopted, how monopoly power is challenged and what new ideas are ultimately released. As we do not yet know, I cannot say what the impact will be with any degree of certainty. |
David Cobham
Heriot Watt University |
Increase (to 4-6%) | Confident |
Assuming no overwhelming adverse shocks! |
Sir Charles Bean
London School of Economics |
Increase (to 4-6%) | Not confident | |
Roger Farmer
University of Warwick |
Increase (to 4-6%) | Confident | |
Matthias Doepke
London School of Economics |
Increase (to 4-6%) | Not confident | |
Jorge Braga de ...
Nova School of Business and Economics, Lisbon |
Increase (to 4-6%) | Confident |
At the 2017 NBER conference on The Economics of Artificial Intelligence: An Agenda Ajay K. Agrawal, Joshua Gans, and Avi Goldfarb it was not yet clear that machine learning would become a general purpose technology but this seems to have happened and not just in developed countries. Assuming the 4% inflation going forward, a global growth above that rate presumes a lessening of geopolitical tensions which at the moment is not apparent. I still bet on the upside rather then no change or fall. |
Evi Pappa
European University institute |
Remain unchanged | Not confident | |
Nicola Fuchs-Sc...
Goethe University Frankfurt |
Increase (to 4-6%) | Not confident |
The impact of AI on politics and society at large could be a threat to growth. |
Gino A. Gancia
Queen Mary University of London |
Increase (to 4-6%) | Confident | |
Michael Wickens
Cardiff Business School & University of York |
Remain unchanged | Not confident |
My guess is that it will increase the rate of growth, but by less than 4% |
Richard Portes
London Business School and CEPR |
Increase (to 4-6%) | Not confident | |
Alessandra Bonf...
Queen Mary University of London |
Increase (to 4-6%) | Confident | |
Ricardo Reis
London School of Economics |
Increase (to 4-6%) | Not confident |
Forecasting future growth over a decade is very hard, so "not confident" is the most relevant part of this answer. |
Agnès Bénassy-Quéré
University Paris 1 & Paris School of Economics |
Remain unchanged | Not confident | |
Paul De Grauwe
London School of Economics |
Remain unchanged | Confident | |
Ugo Panizza
The Graduate Institute, Geneva (HEID) |
Increase (to 4-6%) | Confident |
I am confident because I think that AI will lead to an increase in productivity and this higher economic growth. However, I am not very confident because it is possible that AI could increase unemployment and inequality and this may have backlashes on productivity and growth. |
Robert Kollmann
Université Libre de Bruxelles |
Increase (to 4-6%) | Not confident |
The trend growth rate of global GDP has been remarkably stable since the mid-1970s, despite major technological advances (ICT revolution) and structural changes (e.g., collapse of communism, trade liberalization). It seems unlikely that the long-term trend growth rate of world GDP will be greatly affected by AI. But in the short to medium term (i.e., over the next decade), global growth could increase slightly, say by 0.5%, due to AI. |
Andrea Ferrero
University of Oxford |
Remain unchanged | Not confident at all |
I expect AI to have a significant impact on the economy, but I'm not really sure how. I can imagine that some sectors will benefit more than others, and some may even suffer. The implications for economic growth overall are highly uncertain in my view. |
Cédric Tille
The Graduate Institute, Geneva |
Increase (to 4-6%) | Confident | |
Natalie Chen
University of Warwick |
Increase (to 4-6%) | Confident | |
Philip Jung
University of Dortmund |
Remain unchanged | Not confident at all | |
Wouter Den Haan
London School of Economics |
Remain unchanged | Not confident |
I would not be surprised if there is a substantial level effect and a nontrivial growth effect. But substantially higher sustained growth would surprise me. |
Costas Milas
University of Liverpool |
Remain unchanged | Confident |
I would say: increase between 1% and 2% |
Maria Demertzis
Bruegel |
Increase (to 4-6%) | Confident |
I view the recent developments very favourably. I am not sure whether the impact will be as hight as 4-6% but I certainly think it will be positive |
Linda Yueh
London Business School |
Increase (to 4-6%) | Not confident | |
Lucio Sarno
Cambridge University |
Increase (to 4-6%) | Confident |
| Participant | Answer | Confidence level | Comment |
|---|---|---|---|
|
Ethan Ilzetzki
London School of Economics |
Increase | Not confident | |
|
Łukasz Rachel
UCL |
Remain unchanged | Not confident |
Again, difficult to be confident. I am basing my answer on my reading of historical experience, which is that the displacement effect of technology has put pressure on wage growth but has not led to sustained rise in unemployment in the past. |
Volker Wieland
Goethe University Frankfurt and IMFS |
Decrease | Not confident |
Generally, I would expect increasing scarcity of labour due to demographic effects leading to a reduction in unemployment. This will very likely be the dominating factor. Ceteris paribus I think AI may well decrease unemployment a bit more, even while decreasing hours. |
|
Jagjit Chadha
National Institute of Economic and Social Research |
Remain unchanged | Confident |
it may change the composition of employment but I see no reason why it will over the long run lead to pools of unemployment, unless we respond badly with the requirements for skills in the new-new economy - which I cannot rule out! |
|
David Cobham
Heriot Watt University |
Remain unchanged | Not confident |
It might take longer for the reinstatement effects to balance the other effects. |
|
Sir Charles Bean
London School of Economics |
Remain unchanged | Confident |
While I do not expect AI to have much effect on the overall level of unemployment, I do expect it to lead to substantial changes in the nature of jobs, with an increase in both job destruction and job creation. And there will be an important role for policy to assist those who find their jobs disappearing retrain and find new jobs. |
|
Roger Farmer
University of Warwick |
Remain unchanged | Confident | |
|
Matthias Doepke
London School of Economics |
Remain unchanged | Not confident | |
|
Jorge Braga de ...
Nova School of Business and Economics, Lisbon |
Increase | Confident |
The AI revolution may not cause job polarisation in high income countries as it will affect low-skilled, middle-skilled and high-skilled jobs. However unemployment is likely to increase in developing countries because geopolitical uncertainty involves countries like China, India, Brazil, Nigeria etc and that will have knock-on effects on high income countries. |
|
Evi Pappa
European University institute |
Remain unchanged | Confident |
AI might induce a change in working hours and habits, not necessarily a change in employed workers and possibly a change in the wage paid and productivity (See Erik Brynjolfsson, Danielle Li & Lindsey R. Raymond, NBER WP 31161). |
|
Nicola Fuchs-Sc...
Goethe University Frankfurt |
Increase | Not confident |
Again, the direct effect of AI might rather be positive, but the political uncertainty increases. |
|
Gino A. Gancia
Queen Mary University of London |
Remain unchanged | Not confident | |
|
Michael Wickens
Cardiff Business School & University of York |
Remain unchanged | Confident |
The number of jobs and the level unemployment will not change but the number of hours of work will fall and leisure increase. This is what has happened following past tech improvements. The same thing will happen with AI. |
|
Richard Portes
London Business School and CEPR |
Remain unchanged | Confident | |
|
Alessandra Bonf...
Queen Mary University of London |
Decrease | Not confident | |
|
Ricardo Reis
London School of Economics |
Remain unchanged | Confident |
Historically, the correlation between new technology adoption and unemployment, at the macro level, is close to zero once enough time has gone by. But, a decade is sufficiently short as a period of time, that it could go either way. |
|
Agnès Bénassy-Quéré
University Paris 1 & Paris School of Economics |
Remain unchanged | Not confident | |
|
Paul De Grauwe
London School of Economics |
No answer | Not confident at all |
The option that the employment effect is uncertain is missing here |
|
Ugo Panizza
The Graduate Institute, Geneva (HEID) |
Increase | Not confident |
This is very hard to predict, but I suspect that structural unemployment might increase because there will be a substantial number of workers with skills which are no longer useful or valuable |
|
Robert Kollmann
Université Libre de Bruxelles |
Remain unchanged | Not confident |
Will AI have a major effect on the market frictions and rigidities that determine trend unemployment? Probably not. |
|
Andrea Ferrero
University of Oxford |
Remain unchanged | Not confident at all |
I expect unemployment to increase in the short run as human employment in some tasks and jobs becomes obsolete. In the medium run, the supply will adjust and it's possible that AI may even reduce the natural rate of unemployment in the long run. I'm just not sure. |
Jürgen von Hagen
Universität Bonn |
Remain unchanged | Very confident | |
|
Cédric Tille
The Graduate Institute, Geneva |
Remain unchanged | Confident |
Unemployment effects may be limited, but the impact on income inequality and need for redistribution policy may be large. |
|
Natalie Chen
University of Warwick |
Increase | Not confident | |
|
Philip Jung
University of Dortmund |
Remain unchanged | Not confident at all | |
|
Wouter Den Haan
London School of Economics |
Increase | Not confident |
My concern is that AI may be bad for the more vulnerable in the labour markets like the ones who will not be that easy to adapt to a new environment. |
|
Costas Milas
University of Liverpool |
Increase | Confident |
Increase slightly. |
|
Maria Demertzis
Bruegel |
Remain unchanged | Confident |
There will have to be reskilling. The quicker this happens the less the impact on unemployment. |
|
Linda Yueh
London Business School |
Remain unchanged | Not confident | |
|
Lucio Sarno
Cambridge University |
Increase | Not confident |