After nearly a year of researching the effects of increasing the EU emissions reductions target from 20 to 30%, what are your findings on the implications for jobs and growth?
Probably the most important is that it is time we stopped looking at climate policies in terms of costs and burden sharing. If we organise things like that we'll get nowhere. It's time to look at it in terms of shared opportunities. In Europe, the opportunities are due to the fact that we have a lot of unemployed people and a very low rate of investment. Therefore we have a very low rate of innovation.
We can improve this because climate policy requires massive investment. You need to retrofit buildings. You need to build new power plants and grids. You need new transport systems. These are huge investments and that is good because if we invest more, we can have higher growth, employment and innovation.
The first finding in technical terms is that, if it is done well, we can reduce European emissions by 30% in Europe while increasing our growth rate. In one of the richer regions in Germany, Baden-Wurttemberg, the government has proposed to unilaterally reduce emissions by 30% by 2020 and they're not afraid that this will damage the regional economy at all. Quite the opposite!
This region which, coincidentally, is where the energy commissioner comes from, already has stringent and successful laws for retro-fitting dwellings and they have created networks of industries, which have exchanged experiences of energy savings. These industries managed to reduce energy costs at twice the rate of others.
We can trigger the investments that are needed for climate policies in such a way as to accelerate investment and thereby growth. The kinds of economic models that were used before the economic crisis [of 2008] could not show these opportunities because they were built in such a way that there was only one economic equilibrium, the one we were in.
The financial crisis has shown that there was something problematic about these models. They were OK so long as you only looked at very small changes in the economy, but with large changes they weren't helpful. Reducing emissions and decarbonising the economy is, for sure, a large economic change, so we need models that can represent different equilibriums and transitions.
What are your headline results?
If the policies that we advocate are implemented, you increase the rate of investment from 18% to 22% of GDP. European economic growth can increase by 0.6% per year and over the next decade that will be about 6% more growth. So you can reduce unemployment from an average of about 7.5% to 5%.
It is very important that you do not have an imbalance between old and new member states. The new member states are growing faster than the old ones – fortunately because they are less wealthy – and the gain from such a policy would be evenly distributed between new and old member states.
Which sectors of society did you find would benefit from a move from 20% to 30% and which would lose out?
This depends on how we engineer the emissions reductions. If this is not framed and centred on higher growth investment and innovation rates, then we're in trouble not only with climate policy but with the whole European project. It's high time we developed a bit of confidence otherwise the euro crisis will not have been the last word.
Assuming that we can do it, the sector which will gain most in absolute terms is construction, because you need to retrofit lots of buildings. There is also a need for a lot of construction work for infrastructure. You generate more employment in construction than in the renewables sector.
In relative terms the renewables sector which will gain most because it's so small. But a big relative gain for the renewables sector is still smaller in euros than a small relative gain in the construction sector.
If it is done well, there is no need for anybody to lose out. Reducing emissions in the next 10 years does mean using less coal. You may in the longer term think about carbon capture and sequestration in order to use coal without emitting carbon dioxide but there is no way of doing this on a large scale in the coming decade. Clearly, the companies that are relying on coal will have to switch. But this does not mean that the energy sector loses out because it can also sell renewables and you can switch from coal to gas.
So the energy sector can in fact gain and increase energy prices. This is good as long as income increases faster than energy prices. If my electricity bill increases by one euro and my income by two euros then I'm fine.
I should add that there is a whole array of sectors which are not usually seen in the context of climate policy that become very important. Nanotechnology and robotics are two of them because we'll need new insulation materials and new kinds of windows – very mundane things – but nanotechnology and robotics, including 3D printing, will play a key role in bringing costs down. To a very large extent, this will be a hi-tech operation.
What would be the economic and climate effects of acting later rather than sooner?
The truth is that Europe is not so important anymore and its importance is decreasing. Our role is to be pioneers in engineering and technology and designing new solutions that can be imitated and imported and picked up by the Chinese, Indians and Brazilians.
If we start later in Europe, the most likely thing is that others will take over – you already see this in some fields of renewable energy, solar in particular, where China is already catching up very quickly. If we don't start moving quickly ourselves, then we'll just start watching.
On the question of the climate, I would add that we are too late anyway. It's like nuclear disarmament. We should have disarmed earlier but that is not a good reason to delay things further.
The big risk for Europe is that we miss our huge opportunities and enter a disappointing path of economic development with low innovation rates, low growth and therefore decreasing trust in the whole European project, in particular the euro as a monetary institution.
What obstacles do you anticipate to the measures you advocate being implemented?
I would expect these in the second round, which is implementing these decisions. But obstacles in the first round would involve lobbying by industries that use a lot of energy – steel, cement, aluminium.
These sorts of people believe, perhaps wrongly, that they are threatened and instead of looking specifically at the problems for the European steel industry for instance, and what can be done about them - because its perfectly possible to carefully analyse these - there's a tendency to say 'this is dangerous for everyone, don't touch it!'
The second problem is the mistrust that new member states have against old ones for strong historical reasons, which I understand but which have nothing to do with these problems. Unfortunately the climate of mistrust is worsening because of all the tensions around the euro. The fact that we don't operate in a spirit of high trust and cooperative team play right now is certainly a factor that must be taken very seriously, but we'd better tackle this soon anyway.
What else can you tell us about the modelling, computer simulations and real-world data you used to come to your conclusions?
We started with one of the most sophisticated and tested models currently available: the GEM-E3. It is the standard model used by the EU to assess climate and energy policy. Its model represents the whole world and within that, each European country and within that, more than two dozen economic sectors as well as the households. It has financial sectors, services, different kinds of energy. It's definitely as good as it gets these days, and it embodies what is currently available in terms of data and these economic aspects.
We then tried to learn the lessons from the financial crisis for climate policy and they come as follows:
Number one: economic development is incredibly dependent on expectations. With the financial crisis, it was one bank expecting another bank not to pay back the credit, so you don't give the credit, therefore the other guy goes bankrupt and the real economy gets slowed down because of a breakdown in expectations.
So the first thing we did was to explicitly model shifts in expectations. If we get more positive growth expectations in Europe we can actually realise them. But we have to move together. It's not like Volkswagen has huge growth expectations and no-one else shares them because then no-one will buy their cars.
Secondly, there is strong evidence that by simply doing new things you get better at doing them. For instance, Airbus learned how to produce planes better and better by producing them. If we had not built Airbus in Europe, we would not have had the skills to build planes. You have to actually build things to get better at building them. That's what's called ‘learning by doing'. If you do more you learn more.
We consciously underestimated what was possible from learning by doing, so we took the economy as a whole and not the famous stars like mobile phones and laptop computers where you have much higher rates of learning by doing. You can argue that you should use those high rates of learning in the renewables and nanotechnology sectors, but this we did not do. We stayed on the conservative side.
The third thing is a bit more technical. After a certain number of years of unemployment, people don't get hired any more because employers - often rightly - assume that they would require a lot of re-training. If you reduce unemployment, the separation between insiders and outsiders in the labour market gets weaker so the fraction of people who stay unemployed for very long periods is reduced.
We introduced these three effects into the existing models. All are well-documented empirically, are reasonably well understood theoretically but they all have major technical difficulties. We're actually proud to have started doing this.
Do you know what the GEM-E3's margin of error is?
It depends on what errors you're talking about because these models come with hundreds of numbers. The idea of a margin of error in general is probably not applicable, you have to be much more specific.
If you want to give an economic growth forecast over 10 years then you must accept that you may be wrong by something like a full percentage point over a decade. If you get it wrong by 2% then you're probably in trouble, but if you study the influence of a factor like the carbon price on the growth rate, you can be much more precise. It's not absurd to believe that you can forecast a shift in the growth rate of perhaps a third of a percentage point over a decade. It depends on what you want to look at.
It is difficult because policymakers like us scientists to come with a pseudo-precision that some of us are happy to deliver because it makes you feel good. But we have to distinguish what we can say with great confidence from what we don't really know.
I don't know for sure the growth rates of the European economies over the coming decade but I do know that there's an opportunity to increase the growth rate by something like half a percentage point a year.
It is difficult to make these sort of predictions because so many factors come into play – energy reserve depletion, future energy prices, economic growth, demographics, government incentives, raw material costs, technological processes, etc. – that is why your methodology is so important.
Can you say anything about how many man hours of computers were involved, or even how many computers actually were involved?
In terms of computers, we worked with the facilities in the National Technical University of Athens where this model resides. We combined this with other computations done at the Potsdam Institute of Climate Impact Research, where I work, which is one of the most widely-recognised institutes in the field. Our team was composed of people from all these educational institutes and from Oxford University too. It was coordinated by the European Climate Forum, as well as major companies and NGOs.
We have fantastic computing facilities and we've had work done at the Sorbonne in Paris. We started working on the substance of this last spring, although the formal contract started much later. We were only able to do the work because the financial crisis had led us to work on these issues anyway.
We claim to be more realistic than the models that were done before the financial crisis. We think that crisis is really a major challenge to economic modelling. We learned from it but there is more to be learned. We think that we have got the science right and that's no minor feat. So we are willing to claim with great assertiveness that we can in fact do climate policy as a win-win strategy. We need not do it in such a way that European economies get bogged down. It's not only about computing. It's about judgement.