Environmental impact assessment: is it efficient?

For anyone who's still interested, some more thoughts on whether it's worthwhile to carry out environmental impact assessment for large projects that make a relatively minor contribution to a large problem...

Following last week's discussion on EIA, I think it’s useful to look at the economics of EIA, that is, compare the economic costs and benefits of assessing projects, granting or refusing consent and imposing conditions.

The costs are:

• the cost of the assessment process: the cost to the project proponent of engaging consultants to do an initial assessment of the environmental impact and prepare an environmental impact assessment; the cost to government of considering the impacts, making its own enquiries where necessary and making a decision about consent and conditions; and the cost to third parties of informing themselves about impacts and making submissions to government.
  
• the cost to the proponent of the project being refused or the cost of taking extra measures to comply with conditions of consent.

The benefits are the reduction in external environmental costs imposed on the community by the proposed development. Those benefits arise either because (in extreme cases) the development is refused consent or because conditions are imposed which ameliorate negative impacts of the development.

EIA is therefore efficient from a cost-benefit point of view when the negative externalities that are removed because of conditions (or consent being refused) are greater than the cost to the proponent of complying with the conditions plus the cost to government, the proponent and other interested parties of going through the EIA process.

The benefits are therefore likely to outweigh the costs when the negative impacts of the development are easily identified and can be reduced at a reasonable cost. The EIA process for the bald hills windfarm (see my earlier post) was a classic case where the EIA process could have been efficient but where the decision in my opinion was not. The relevant environmental impacts were the impacts on the orange-bellied parrot and could be estimated as the death of one bird on average every 109 years. I don’t know the cost of the assessment process but the cost of the decision was very substantial: the Minister refused consent to a substantial and otherwise commercially viable energy development. Whether the benefits of that decision outweigh the costs can be debated but, as Harry Clarke has pointed out, the costs of the decision are certainly far higher than the costs of imposing conditions on the development that could have had a far greater benefit for the survival of the parrot: the parrot can be bred in captivity and returned successfully into the wild at a quite modest cost: one that the windfarm could certainly have afforded. This measure could quite cheaply have benefited the parrot population by two or three individuals per year. Compare that to an expensive decision that benefits the population by, on average, one individual every 109 years.

When the environmental impact that you’re worried about is greenhouse emissions from downstream use of a product, I doubt that EIA is an efficient way to attempt to reduce those impacts. First, as discussed last time, the impacts from this particular development are difficult to assess so the assessment process is likely to be difficult and costly. Secondly, the conditions that can be imposed are unlikely to deal with the problem. The only substantive way to reduce emissions from burning the coal from a mine is not to mine it (other than – perhaps - placing some conditions on how it’s processed to, say, reduce the fuel’s moisture content). So the only effective decision the government can make is to approve or not approve the mine. Conditions about how the fuel is used etc really need to be placed on the user of the fuel. As discussed yesterday, since power stations can source their coal from anywhere, it is questionable whether refusing consent for a mine would reduce greenhouse emissions in any case.

Cumulative impacts

A closely related issue is that of ‘cumulative impacts’. This is the ‘death by a thousand cuts’ problem: each development has only a small, possibly trivial, impact on its own but collectively the impact is serious. Assessment of each project on its own fails to protect us from these serious cumulative impacts.

The solution, according to the applicants in the Isaac Plains case, is that the ‘cumulative impacts’ contributed to by a proposed development should be considered by the decision-maker. As I’ve discussed before, this is the basis on which the federal Environment Minister made a decision earlier this year to refuse consent to a windfarm: it would have an impact on the orange-bellied parrot that is trivial on its own. But the impact, in combination with existing and under-construction windfarms, was so serious that the proposed development should be refused: enough was enough and the Minister had to draw a line somewhere.

I’m not sure about this. In principle, a project should be assessed on its own merits. You should compare the costs of this project with the benefits of this project and determine whether the project should go ahead or what conditions should be attached. It seems inappropriate to compare the costs of this project and others like it with the benefits of this project and others like it and then determine whether the project should go ahead or what conditions should be attached – it just seems to be asking the wrong question.

That said, the alternative is not particularly attractive either. Consider a threatened plant affected by development. There are 100 hectares of the plant left. A development proposes to clear 1 hectare of it for some modestly beneficial community infrastructure. The modest community benefits are deemed to outweigh the environmental costs (it’s only one hectare) so the development is approved. Similar developments are proposed again and again and the equation is the same. Imagine that soon there’s only 30 hectares left. Suddenly a project that will bring enormous benefits to the community is proposed but it will involve clearing 10 hectares: a third of the entire distribution of a now critically endangered species. The choice is now difficult: miss out on huge community benefits or push a species to the brink of extinction. Perhaps the choice could have been avoided if, earlier on when 100 hectares were still left, a decision had been made that the cumulative pressures on this plant were likely to be huge over the next few years and so only projects that had either very large benefits or negligible impacts on the plant would be considered.

It is difficult to do this when approving individual projects though. At which stage do you decide that although the benefits of this project exceed the costs of this project, you’ll nevertheless reject the project because, if projects like this keeping being approved, in a couple of years we’ll reach a stage where it’s too costly to approve any more?

The cumulative impacts problem is a big one and EIA is one way of dealing with it. A better way in my opinion is to deal with it at more strategically: either through planning law (eg, zoning certain areas to protect threatened species so that these conflicts don’t emerge) or economic instruments (providing economic incentives to protect and rehabilitate threatened species sites).

And those instruments are a discussion for another day...