Critique of willingness-to-pay and willingness-to-receive methods for valuing natural capital

One valid criticism of economic valuation approaches is where they are based on willingness-to-pay (to acquire the asset not currently owned) or willingness-to-accept (to accept compensation for the loss of the asset currently owned).  In many cases, and especially in relation to valuing nature, there is likely to be an element of hypothetical bias in people’s answers to surveys of their willingness-to-pay or willingness-to-accept preferences. 
 
Hypothetical bias can enter the picture because the respondent is being confronted by a contrived, rather than an actual, set of choices. Since he or she will not actually have to pay the estimated value, and has potentially very little understanding of the significance of the hypothetical scenario underlying the question, the respondent may treat the survey casually, providing ill-considered answers.
 
A particular case of this is where the respondent is being asked about the value of a natural asset (eg natural capital comprising part of the global commons), when it is currently valued at zero and the respondent (via the global industrial supply chains) is currently not paying anything for the use (or abuse and exploitation) of that asset.
 
These sorts of valuation methods can also be undermined by the personal circumstances of the respondent.  A respondent with very small amounts of financial resources (eg income) is likely to give much smaller willingness-to-pay answers than one with much larger financial resources.
 
Let me paint a hypothetical scenario to illustrate this.  For those of you who have heard of the Biosphere 2 experiment in the 1990s, imagine a re-run of that 2-year experiment. For those unfamiliar with it, Biosphere 2 was a project where 7 volunteers lived inside a closed set of connecting bio-domes containing air, water, soil, plants etc with no inputs or outputs (wastes) allowed to enter or leave the domes, except for the sunlight falling on them and the energy ‘leakage’ out through the glass walls at night.   Everything was recycled that possibly could be.
 
In the imaginary re-run, which I’ll call Biosphere 3, because of the strict limits on availability of such materials, suppose that each inhabitant had a daily budget of financial resource, and each of the main materials required for survival was made exclusive to each “consumer” (eg with breathable air provided via breathing apparatus and personal cylinders, and exhaled air collected in similar cylinders for recycling, water provided in personalised bottles, food in personalised containers) But suppose further that this was all provided as a public good for the first six months, on a strictly need-to-consume-for-survival basis.  Suppose also that some inhabitants had financial resources very different from others – say, some had $100 per day while others had $100,000 per day.    
 
In this scenario, what would willingness-to-pay or willingness-to-receive questioning and valuation methods reveal about the values of the various survival essentials?  I would guess that the answers would be very different between participants with different financial resources, and they would also be very different with passage of time.
 
On day one, probably everyone would be willing to pay almost all, if not actually 100%, of their ‘budget’ for breathable air (even if they don’t actually have to pay for it), on the basis that none of them could survive the day without breathable air, whereas everyone could survive without water or food for one day.  The trouble with that is that it gives a value for breathable air of $100 for some respondents, $100,000 for other respondents.  As long as this doesn’t affect the actual amounts of air, water and food provided (because they are being provided as a public good, not as a market-traded exchange) there is no problem with continuing provision.
 
The situation would get more complicated some days later, because after some days without water, survival would become a matter of securing water as well as air, and after an important threshold of deprival, no amount of air (or money) would compensate for lack of water, without which death would follow.  Therefore, everyone would be prepared to pay something for some water, even if all the rest of their budget would be allocated to breathable air in their willingness-to-pay answers.  So, perhaps their willingness-to-pay valuations would represent 90% of their “budget” for air and 10% for water each day, in order to survive until the next day.  Some time after that point, edible food would have to be part of their answers, to avoid theoretical death, on the basis that no amount of money, air and water can stop someone dying if they don’t have food.  So perhaps their valuations of willingness-to-pay would represent 90% of the budget allocated to air, 7% to water and 3% to food each day, in order to survive to the next day.  And so on.  Perhaps this situation and the associated willingness-to-pay values would reach some form of stable equilibrium after, say, six months.
 
It can easily be seen that the “values” of each of the necessities, as measured by willingness-to-pay, would vary considerably between participants, and would vary considerably over time during those first six months.
 
Now consider the scenario that the actual costs of providing all those necessities each and every day in Biosphere 3 was a lot higher than any of the personal financial resources of the ‘poorest’ of the individual participants, but not beyond the financial resources of the ‘richest’ participants.  Assume, for example, that the cost of breathable air was $50,000 per day per person, water was $30,000 and food was $10,000.
 
Further, assume that the “provider” after the first six months changed, and was then a computer or Artificial Intelligence operating on market exchange principles (a complete contrast from providing essentials as a public good), and was unwilling (or unable) to compromise and incur a ‘loss’ on provision (eg the provider was not going to accept a lower price than the actual cost of provision).  It will be clear that this is a very different state of affairs compared with the first six months of public good service provision.
 
The willingness-to-pay answers for the poorest participants under this new model for ‘traded’ provision would now be, in fact, an irrelevant and gross under-estimate of the value of the provision, for any answers they give, even if they answer 100% of their personal budget (ie $100 per day) as the willingness-to-pay value for any one of the essentials.    They simply won’t survive day one, because they don’t have sufficient financial resources to pay for the true cost of the provision of the essentials for life.  Their survival during the first six months was dependant on the provider subsidising the true ‘cost of living’ (ie incurring the cost of the public goods themselves and not passing them on to the consumer).
 
At least some of the “richer” participants might well survive the new regime for provision, but even though some of them might be able to cover the average actual costs of provision, they might not all survive the whole 2 years.
 
For example, if there is a bidding war, and if the Artificial Intelligence has profit maximisation as its aim, its optimum market strategy might be to use its monopoly power to manipulate prices to extract the maximum revenue, even if this means only very few participants surviving to the very end of the 2 years.
 
This might seem a very extreme scenario, contrived in order to challenge the validity and accuracy of willingness-to-pay or willingness-to-receive valuation methods.  However, if you consider for a moment the overall insights derived from Kenneth Boulding’s “Spaceship Earth” conceptual writings, it doesn’t take a huge leap of the imagination to see how the whole of the Earth can be thought of as a larger version of Biosphere 3.  In fact, the reason the 1990’s dome experiment was called Biosphere 2 was because it’s designers considered the Earth to be “Biosphere 1”.  
 
So, Biosphere 1 (the current Earth we live on) has some similarities with Biosphere 3.  The Earth is a finite, closed system, except for the sunlight falling on it (and the solar energy escaping through the atmosphere into space). The main differences between the Earth (Biosphere 1) and my hypothetical, imaginary Biosphere 3 are matters of scale, the key ones being the overall size of the Earth and its biological and chemical systems, and the size of the global human population.
 
With this perspective in mind, my “Biosphere 3” thought experiment can be seen to challenge the willingness-to-pay and willingness-to-receive valuation methods used in Biosphere 1 (the Earth we live in).
 
The Biosphere 3 scenario highlights the downsides of a marginalist approach to valuation.  Willingness-to-pay and willingness-to-receive approaches are difficult to apply without the respondents falling into a marginalist perspective.  That is, they are very likely (even if subconsciously) to consider a price or value in a way that is heavily influenced by their current circumstances and by the existing state (and prices / values) of the objects in the existing systems of supply (whether market-based or public goods provision).  Many of these existing prices / values are effectively zero or heavily subsidised in current global supply chains, so this is likely to give a very strong bias towards very low valuations resulting from these methods of valuation.