Wednesday, August 8, 2012


Walter Westman Lecture on Science, Humanity and the Environment International House, University of Sydney, Wednesday, August 8, 2012

I was pleased to attend the first of these Walter Westman lectures, in 2008, given by my friend Steve Hatfield-Dodds, and I’m honoured to deliver this year’s lecture. I want to talk about economic growth - a topic of concern to many people who think deeply about science, humanity and the environment - and specifically about whether it would be feasible for us to move to a ‘steady-state’ economy.

Virtually all our business people, economists and politicians believe it to be not just possible but desirable for the economy to continue growing forever - and the faster the better. They regard the achievement of growth as one of the most important objectives of government, and they have no doubt this is in accord with the wishes of almost everyone in the electorate.

Why do they regard economic growth as such a good thing? Because when the economy’s generation of income - and its production of goods and services, which is the same thing, and is conventionally measured by the growth in real gross domestic product - grows faster than the population is growing, income per person is rising, meaning that, on average, our material standard of living is rising. Doesn’t that sound good to you?

But the advocates of growth believe it brings with it many other advantages. They argue that the richer we get, the more easily we can afford to spend money on fixing the environment. They argue that rising real incomes should also directly benefit the poor and, as well, that when incomes are rising it’s easier to get agreement to redistributing income in favour of the poor. They further argue that growth in the economy is necessary to create the additional jobs needed to provide employment for a growing population of working age.

So why do an increasing minority of people oppose continuing economic growth? Because of their belief that unending economic growth is ecologically unsustainable and, indeed, physically impossible. They see the global economy as a sub-system of the global ecosystem, which is of fixed size. If the ecosystem can’t grow, then there must be a limit to the extent to which the economy can grow within it. They look at all the damage economic activity has done and is doing to the natural environment - the damage to soil, forests, waterways and fish stocks, the destruction of species and, most obvious and pressing, the emission of greenhouse gases - and they conclude we must be close to the ‘limits to growth’. To press up to those limits or even exceed them must surely damage the natural environment to such an extent that huge, possibly irreparable damage is done to the economy, not just the environment.

That’s the most fundamental, pressing reason for wanting to call a halt to growth, but there’s a supporting reason from the ‘science of happiness’. Psychologists and a few economists studying what they prefer to call ‘subjective wellbeing’ have concluded that once people’s incomes reach a certain fairly modest level, further increases do little or nothing to raise the ‘aggregate happiness’ of the people in a country. This is because we so quickly adapt to any change in our material circumstances, with improvements soon being taken for granted as we aspire to something bigger and better.

But it’s also because, as studies show, what makes everyone but the poor happier as individuals is not an increase in our income along with everyone else’s, but an increase in our income relative to everyone else’s. It’s this that would feed our desire to feel superior to others and also permit us to demonstrate that superiority to the world by means of our superior possessions. The trouble with such aspirations, however, is that they involve a zero-sum game: I can advance my position in the pecking order only at the expense of the positions of all the people I pass. Such status competition - such consumption competitions - is socially wasteful: it uses up a host of scarce economic resources without making any net addition to total happiness. For the managers of the economy, this creates two problems. First, all they can ever do is raise incomes overall; there’s nothing they can do to raise the relative incomes of everyone in the economy. So their growth-favouring policies do little or nothing to increase the community’s subjective wellbeing - which must surely be the justification for their efforts. Second, it means that, for an affluent economy such as ours, most of the annual increase in our real incomes that the economic managers have laboured to produce is dissipated on the socially wasteful purchase of ‘positional goods’ - goods and services whose purpose is to demonstrate to others our superior position in the social order. For me, this is a powerful reason for not being too dismayed by ecologists’ insistence that we can no longer afford economic growth: if so, we won’t be giving up all that much.

This brings me directly to my topic: is a steady-state economy - and economy that doesn’t grow - feasible? First I’ll examine the reasons people argue it isn’t feasible, then I’ll examine what we’d need to do to make it feasible.

The growth imperative is so deeply ingrained in our thinking, so much an assumption underlying so much of what we say - even what I say - that many people imagine the economy is like a bicycle: if you stop going forward you lose your balance and fall off. Fortunately, the analogy isn’t apt. Though many people would have to adjust their thinking in a steady-state economy, it wouldn’t collapse just because it wasn’t growing. What would cause deep problems, however, is if the economy was steadily shrinking, particularly if prices were falling.

There’s obvious truth to the argument that when incomes are growing it’s easier to afford to repair the environment. But that argument becomes dubious when we reach the point where the growth itself is adding to the environmental damage. We have to destroy the environment to afford to save it? It’s hard to imagine how the environment could end up ahead on that deal.

There’s more truth to the argument that when incomes are growing it’s easier to give the poor a better deal, including by redistributing income from the better-off to the less well-off. It’s factually correct - in Australia, though not in America - that real growth in national income over the years has led to real growth in the incomes of households at the bottom of the distribution, as well as in the middle and at the top. What’s also true, however, is that incomes at the very top have been growing much faster than all other incomes. I think advocates of a steady-state economy have to accept that, yes, the absence of growth would increase the political resistance to greater redistribution in favour of those at the bottom. But just because growth makes greater redistribution easier doesn’t necessarily mean redistribution happens. It hasn’t happened sufficiently to prevent the gap between rich and poor widening significantly over the past 30 years or so, notwithstanding all the growth we’ve enjoyed.

The strongest anti-steady-state argument is that we need economic growth to provide employment for our growing population of working age. It’s pretty much the only argument I get from the ecologically aware economists I talk to. But I don’t think it’s insurmountable. The first reservation is that, were it not for immigration, our working-age population wouldn’t be growing (as is the case for most of the advanced economies and will soon be the case for China). We could adjust our net migration to keep the working-age population steady. The second reservation is that, even if our working-age population was growing, we could respond to the problem by job-sharing. Here I’m not only referring to the idea of two or more part-time workers sharing the one full-time job, but to the more fundamental solution that rather than continuing to take the continuing improvement in the productivity of labour in the form of ever-higher real wages, we could do what the futurists of the 1960s and 70s expected we’d do and take it in the form of shorter working hours.

Before I turn to the more positive question of how we’d go about achieving a steady-state economy, we should clarify an issue I probably should have raised much earlier. In all that follows I’ll be leaning heavily on the leading thinker in the area of steady-state economics, Professor Herman Daly of the University of Maryland. There’s enormous terminological confusion between scientists and economists on what exactly they mean by the word ‘growth’. Scientists take it to mean something very different from what economists do, which means much of what little debate passes between them flies over the heads of the other side. I’m sure the ground of disagreement between them would be greatly reduced if only this terminological confusion could be ended.

What ecologists want is an end to growth in the ‘throughput’ of natural resources. If you think of the economy as a machine, we put inputs in one end of the machine, and take outputs out of the other end. To an ecologist, the inputs of concern to them are natural resources and ‘ecosystem services’; the outputs of concern to them are an equivalent amount of waste - in the form of landfill, sewage and all the many types of pollution, including greenhouse gases. In conformity with the laws of thermodynamics, the ecologists worry as much about the emission of waste - and the ecosystem’s ability to absorb that waste - as they do about the using up of natural resources. This is why what they seek is an end to growth in the throughput of such resources. I think many of them imagine this would be achieved if GDP ceased to grow.

But the economists conceptualise things very differently. To them, the inputs to the economic machine aren’t just natural resources, but also the other economic resources: labour and capital - physical capital in the form of machines, structures and infrastructure. (The input from ecosystem services is ignored.) To their eyes, the output from the economic machine isn’t waste (it gets ignored) but all manner of goods and services. What real GDP measures is the growth in the output of goods and services over time. (Since those of us who work earn our income from our contribution to the output of goods and services, real GDP also measures the growth in real income.)

So what is it that causes GDP - output of goods and services - to grow? Two things. First, any increase in the throughput of economic resources: natural resources, but also labour and capital. But, second - and this is the bit that goes straight over the heads of most ecologists - any increase in the efficiency of the economic machine at turning inputs into outputs. Economists call this ‘productivity’, which they define as output per unit of input. The productivity of the economic machine increases almost continuously each year, and has done since the start of the industrial revolution. What causes ‘multi-factor’ productivity to improve is the continuing pursuit of economies of scale, the increasing specialisation of labour, the rising knowledge and skill of the workforce, and technological advance: the invention of better machines and better ways of doing things. Now get this: over the long term, productivity improvement accounts for the lion’s share of our rising real income per person and our rising material standard of living.

The point is that when economists hear people say they want an end to growth, they assume that means they want an end to productivity improvement. They find this prospect appalling. But this is not what ecologists want. All they want to stop is growth in the throughput of natural resources - which isn’t something most economists would relish, but isn’t nearly as frightening. And this means GDP could still increase, provided that increase came from improved productivity, not increased use of natural resources.

Clearing up this misunderstanding allows us to envisage more clearly what a steady-state economy would look like. It would be an economy that didn’t get bigger in its impact on the environment - that was ecologically sustainable - but did get better, in terms of the quality of our lives. It would be an economy that didn’t grow, but it wouldn’t be an economy that was stagnant, that never changed. It wouldn’t be an economy where people had to stop striving - to build a better mousetrap, write a symphony or find the cure for cancer. Many economists instinctively fear a steady-state economy would stifle the incentive to innovate. But that fear’s not justified. Indeed, you could argue that, with the quantitative route to improvement blocked off, the qualitative route would gain more attention. Herman Daly’s way of making the distinction is to say economic growth (pushing more resources through a physically larger economy) is bad, but economic development (squeezing more welfare from the same throughput of resources) is fine.

But how would we go about reorganising the economy so that we no longer increased the throughput of natural resources? It wouldn’t be easy, but nor would it be terrifically hard. It actually represents nothing more than a design problem - one the economics profession is well-equipped to solve, should we decide to give it that task. We’d still have a capitalist, market economy where market forces continued to determine economic outcomes and to drive the push for greater efficiency in resource use, within the framework set by government. The big difference would be the government adding a new constraint to the operation of market forces: a limit on the consumption of natural resources.

How would we achieve that limit? By using the same ‘economic instrument’ we’ve already begun using to limit the burning of fossil fuels: a system of tradable permits. You impose a cap on the total quantity of a certain class of natural resource permitted to be consumed in a year, and auction to producers permits to use the resource up to the cap. The more efficient firms are at doing what they want to do while using fewer natural resources to do it, the less they have to spend on permits - thus harnessing market forces to help reduce the use of those resources. Firms that discover they have more permits than they need are able to trade them for money with firms that discover they need more permits than they have. By such means the burden of limiting resource use to the cap is transferred to those firms able to reduce their resource use most cheaply, thereby limiting the loss of income to the community involved in achieving the limit on resource use. As firms became more efficient at reducing their natural resource use - including by the invention of new technological solutions to the problem - it would possible, if desired, to lower the cap and, hence, the quantity of resources used, at no increased cost to the community.

The purpose of such a cap-and-trade scheme would be, of course, to raise the price of natural resources - and the prices of goods with a high natural-resource component - relative to the prices of all other goods and services. In line with the most orthodox economics, it’s this change in relative prices which would motivate producers and consumers to reduce their resource use, and do so with minimum loss of economic efficiency. Economists believe changes in relative prices are very effective in bringing about changes in the behaviour of producers and consumers.

This process would, of course, lead to a once-off increase in the general level of consumer prices, which might be quite a significant increase. Many of you would be concerned about the effect on the cost of living, particularly for pensioners and low income-earners. But, as with our present carbon tax, once the cap-and-trade scheme had brought about the desire changed in relative prices, the proceeds from the sale of permits - analogous to the proceeds from a carbon tax - are available for use to reduce the rates of other taxes - the obvious one being income tax - and increase the rates of pensions and benefits such as the family allowance, thereby compensating households for the increase in their cost of living. So I don’t see a reason to be concerned about the effect of the move on the welfare of low income-earners. Such a re-jig of the tax system would be a classic example of what environmental economists mean when they call for the burden of tax collection to be shifted from taxing ‘goods’ (such as labour and capital) to taxing ‘bads’ (such as greenhouse gas emissions and the consumption of natural resources). You raise the same amount of total tax revenue but, in the process, you discourage activities you want to discourage rather than activities you don’t want to discourage.

Raising the prices of natural resources relative to the prices of other resources - labour and capital - could be expected to have various desirable side-effects. First, it would increase the economic incentive for people to recycle natural resources and repair rather than replace appliances with a high materials-component.

Second, changing the relative prices of economic resources could be expected to change the focus of the private sector’s continuing search for greater efficiency - economising, if you like - in the use of economic resources and, hence, improved productivity. For all the time since the Industrial Revolution, most of the economising effort - including most technological advance - has been, quite logically, directed towards economising in the use of the most expensive resource: labour. But if we were to make natural resources more expensive than labour - particularly if the scheme involved a fall in the main tax on labour, income tax, thereby lowering its effective cost - this should mean a lot more entrepreneurial effort would be directed towards reducing the economy’s despoiling of the natural environment.

There are many more implications of a steady-state economy I could explore, but that’s enough to be going on with. Is a steady-state economy feasible? Yes it is.