Part 2: Valuing environmental effects

W H Davies encouraged us to stand and stare at, and appreciate, the natural world.

Alongside personal and intra-household activity, just discussed, the natural world is of course another area where it is challenging to measure and account for the effect of human behaviour – at least, in so far as it goes beyond simply standing and staring.

As a very familiar example, pollution, and associated environment degradation, is not included as a cost in measures of the output and income associated with the polluting activities – even though economists would regard such “external” costs (or “externalities”) as just as much a subtraction from social value as are the costs borne by private individuals. (Environmental effects are measured in the “satellite” system of Environmental Accounts but not incorporated in the main accounts.)

One of the most damaging, and perhaps the most damaging, form of pollution in modern times is of course the emission of greenhouse gases. It therefore represents a priority case for assessing the value of the effects of pollution on the environment.

However, the challenges of establishing the right principles for the valuation of greenhouse gas emissions are not trivial and may vary depending on the objectives of the exercise.

First, the appropriate boundary for measuring the damages associated with greenhouse gas emissions must be chosen.

The damage that current and future UK-sourced emissions impose on people in the UK itself will be relatively very small – greenhouse gas emissions are a global externality, and the large majority of the damages from UK emissions will be felt by people in other countries. (This is not true, of course, for historic omissions.) At the same time, the vast bulk of the damage that will be imposed on people in the UK arises from other countries’ emissions.

Different approaches to the “boundary” will be appropriate if the purpose is to measure the global costs associated with the greenhouse gas emissions from UK economic activity or the impact on UK living standards from greenhouse gas emissions.

Having selected the geographical boundary for measuring emissions, the valuation of the damage associated with emissions poses further challenges.

Most fundamental may be the ethical position adopted. For example, is valuation to be based only on the impact on existing and future humans, or is the environment to be accorded some intrinsic value?

If the latter position is taken, the basis for valuation is not obvious. If the former position is adopted, one obvious option would be to use carbon values mandated by the UK Government for use in policy appraisal. However, these values are designed to be consistent with the UK Government’s political choice of targets designed to achieve net zero in 2050. The values are regularly updated as new information on costs and past performance becomes available.

The official values are therefore not intended to represent a direct estimate of the damage costs of greenhouse gas emissions, and their use will not deliver estimates that are comparable either across countries or through time.

An alternative would be to assess the value of greenhouse gas emissions based on a direct estimate of damage costs. However, estimates made by academics and researchers are dependent on a range of assumptions, including the sensitivity of the climate to greenhouse gas emissions, and are highly divergent.

A recent review found that estimates of long-run costs of an increase in global temperature of one degree centigrade ranged from 1 to 25 per cent of global annual GDP by the end of the current century. And, in principle, to this wide range should be added a highly uncertain risk premium to take account of possible tipping points and catastrophic outcomes with low but uncertain probability.

A further complication arises from the spatial variability of damage cost estimates across the globe, with estimated costs much lower in colder and temperate countries, and with some studies even showing net benefits in certain dimensions for some such places. Again, the appropriate choice here will reflect the purposes to which the estimates are to be put.

In assessing damage costs, a key issue is the discount rate to be used to adjust costs downwards depending on how far into the future that they occur. The choice of discount rate reflects both theoretical considerations and value judgements, including the weight to be put on the welfare of future generations. Researchers have adopted various approaches, with, at one extreme, the Stern Review arguing that the discount rate for future damage should be close to zero, except in so far as income levels are expected to increase. (The rate is above zero because of the possibility of human species extinction for other reasons.)

While the principle that external costs should be properly accounted for in measuring output and income seems intuitively attractive when faced with the example of the very great damage associated with greenhouse gas emissions, adopting it in this case raises the question of how extensively the principle should be applied.

While greenhouse gases may be the most prominent current example of an external cost, externalities are pervasive across economic activity, with large external benefits, rather than costs, to some activities (such as research and development and health improvement).

While not all externalities raise ethical issues that require value judgments to the same extent as with greenhouse gas emissions, in most cases quantification, and still more monetisation, is very challenging. There is often consensus about the direction of the effects, but, as with greenhouse gas emissions, estimates of scale can vary widely.

As a result, quantification and valuation will in part depend on assumptions which are highly uncertain and where different assumptions would result in widely different valuations. (International guidance on valuation may address concerns about comparability across countries but cannot resolve underlying uncertainties and the associated arbitrariness of assumptions.)

Furthermore, it would appear infeasible, given their pervasive nature, to adopt a general policy of valuing externalities and incorporating them in headline measures of output and income. So, the selection of which to include and which to exclude would also appear inevitably to reflect value judgements and be to some extent arbitrary.

While the results of the valuation of externalities and their incorporation in headline indicators of output and income may be welcomed by some stakeholders, as with household activities, this is not likely to be the case for all (including perhaps those whose responsibility includes macroeconomic management, for whom any economic “signal” may be obscured).

At the start of this think piece, I said I was prompted by WH Davies’ poem to reflect on the value we attach to personal activities and to our impact on the environment. I don’t think I have been able to reach any general conclusions, other than that measuring the value of such activities and impacts raises a big issue of principle and that people can reasonably disagree about the approach that should be taken – and how far it should go. User needs are likely to diverge, and these differing needs will need to be balanced.

However, as I say above, I think that going any distance down the path of measurement in these areas makes major value judgments unavoidable. The results will also reflect a potentially high level of uncertainty about the assumptions appropriate to valuation. In consequence, there is a need for complete, and intelligent, transparency about the value judgements and assumptions made. And I therefore think there are particular grounds for caution about the use of aggregate indicators which incorporate elements requiring multiple difficult value judgments and arbitrary assumptions.